Re: master ff4de1b: Fix quoting style in Lisp comments

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

Hi Eli,

On 12.09.2021 19:51, Eli Zaretskii wrote:
> -;; `eval' is hence the core of the culprit.  It's used on:
> +;; 'eval' is hence the core of the culprit.  It's used on:
> 
> -;; encouraged to use `lisp-data-mode' instead.
> +;; encouraged to use 'lisp-data-mode' instead.
> 
> -;; The time- functions below translate nil to `current-time' and
> -;; accept an integer as of Emacs 25.  `decode-time' and
> -;; `format-time-string' accept nil on Emacs 24 but don't accept an
> +;; The time- functions below translate nil to 'current-time' and
> +;; accept an integer as of Emacs 25.  'decode-time' and
> +;; 'format-time-string' accept nil on Emacs 24 but don't accept an

Is this a good idea?

I'm not sure we ever agreed to standardize on straight quotes in Lisp 
comments (as opposed to plain text files like NEWS).

More importantly, the symbol references inside `...' are font-locked 
with font-lock-constant-face, and references inside '...' do not.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Dmitry Gutov <dgutov@yandex.ru> writes:

>> +;; The time- functions below translate nil to 'current-time' and
>> +;; accept an integer as of Emacs 25.  'decode-time' and
>> +;; 'format-time-string' accept nil on Emacs 24 but don't accept an
>
> Is this a good idea?
>
> I'm not sure we ever agreed to standardize on straight quotes in Lisp
> comments (as opposed to plain text files like NEWS).
>
> More importantly, the symbol references inside `...' are font-locked
> with font-lock-constant-face, and references inside '...' do not.

Yeah, I'm against changing `foo' to 'foo' in comments.  It's just too
confusing to have a different syntax in the doc strings and in the
comments.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: emacs-devel@gnu.org,  Eli Zaretskii <eliz@gnu.org>
> Date: Mon, 13 Sep 2021 10:06:31 +0200
> 
> Yeah, I'm against changing `foo' to 'foo' in comments.  It's just too
> confusing to have a different syntax in the doc strings and in the
> comments.

When we had this discussion in the past, people said the ` character
looks very ugly on today's displays.  That's why we settled on '..'.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Phil Sainty]

On 2021-09-14 00:01, Eli Zaretskii wrote:
> When we had this discussion in the past, people said the ` character
> looks very ugly on today's displays.  That's why we settled on '..'.

FWIW, I never really see those quote characters when reading comments
in code, because the face change makes the quoted text pop out.

I would also vote for considering code comments as more like "code"
than "plain text" documentation for these purposes.


-Phil

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Mon, 13 Sep 2021 01:42:08 +0300
> 
> Hi Eli,
> 
> On 12.09.2021 19:51, Eli Zaretskii wrote:
> > -;; `eval' is hence the core of the culprit.  It's used on:
> > +;; 'eval' is hence the core of the culprit.  It's used on:
> > 
> > -;; encouraged to use `lisp-data-mode' instead.
> > +;; encouraged to use 'lisp-data-mode' instead.
> > 
> > -;; The time- functions below translate nil to `current-time' and
> > -;; accept an integer as of Emacs 25.  `decode-time' and
> > -;; `format-time-string' accept nil on Emacs 24 but don't accept an
> > +;; The time- functions below translate nil to 'current-time' and
> > +;; accept an integer as of Emacs 25.  'decode-time' and
> > +;; 'format-time-string' accept nil on Emacs 24 but don't accept an
> 
> Is this a good idea?

We decided to use this style in plain-text documentation long ago.

> I'm not sure we ever agreed to standardize on straight quotes in Lisp 
> comments (as opposed to plain text files like NEWS).

Comments, NEWS, PROBLEMS, DEBUG, commit log messages (thus the
generated ChangeLog files), and any other form of plain-text
documentation.

> More importantly, the symbol references inside `...' are font-locked 
> with font-lock-constant-face, and references inside '...' do not.

If we want to highlight 'like this' (and I'm not sure we do, but I
won't object if we did), then let's make such a change in font-lock,
by all means.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Eli Zaretskii <eliz@gnu.org> writes:

> We decided to use this style in plain-text documentation long ago.

I don't think comments in .el files count (and I don't recall this
decision being made, but perhaps I wasn't here).

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: Dmitry Gutov <dgutov@yandex.ru>,  emacs-devel@gnu.org
> Date: Mon, 13 Sep 2021 13:41:11 +0200
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > We decided to use this style in plain-text documentation long ago.
> 
> I don't think comments in .el files count

So this is only about *.el?  What about comments in other files, like
*.c and *.h? are they different, and if so, why?

Re: master ff4de1b: Fix quoting style in Lisp comments

[martin rudalics]

 >> We decided to use this style in plain-text documentation long ago.
 >
 > I don't think comments in .el files count (and I don't recall this
 > decision being made,

Neither do I.

 > but perhaps I wasn't here).

What would happen when commenting code out/in?

martin

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

> What would happen when commenting code out/in?

I don't think the `...' vs '...' quoting style is affected by this (not
vice versa).


        Stefan "trying to stay out of this discussion ;-)"

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

Stefan Monnier [2021-09-13 12:53:56] wrote:
>> What would happen when commenting code out/in?
> I don't think the `...' vs '...' quoting style is affected by this (not
> vice versa).

That last "not" was supposed to be "nor".


        Stefan

Re: master ff4de1b: Fix quoting style in Lisp comments

[martin rudalics]

 >>> What would happen when commenting code out/in?
 >> I don't think the `...' vs '...' quoting style is affected by this (not
 >> vice versa).
 >
 > That last "not" was supposed to be "nor".

If the rule is that comments should use '...' rather than `...',
commenting code out/in is affected by that rule.

martin

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

martin rudalics [2021-09-15 11:27:40] wrote:
>>>> What would happen when commenting code out/in?
>>> I don't think the `...' vs '...' quoting style is affected by this (not
>>> vice versa).
>> That last "not" was supposed to be "nor".
> If the rule is that comments should use '...' rather than `...',
> commenting code out/in is affected by that rule.

I can't see why:
- the rule would presumably only apply to actual prose comments, not to
  commented out code.
- code uses neither `...' nor '...' (except for the content of
  docstrings, admittedly).


        Stefan

Re: master ff4de1b: Fix quoting style in Lisp comments

[martin rudalics]

 > I can't see why:
 > - the rule would presumably only apply to actual prose comments, not to
 >    commented out code.

OK.

 > - code uses neither `...' nor '...' (except for the content of
 >    docstrings, admittedly).

The doc-strings are what I had in mind.

martin

Re: master ff4de1b: Fix quoting style in Lisp comments

[Juri Linkov]

>>>> What would happen when commenting code out/in?
>>> I don't think the `...' vs '...' quoting style is affected by this (not
>>> vice versa).
>>
>> That last "not" was supposed to be "nor".
>
> If the rule is that comments should use '...' rather than `...',
> commenting code out/in is affected by that rule.

Why not to use proper Unicode quotation like ‘...’?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Juri Linkov <juri@linkov.net>
> Cc: Stefan Monnier <monnier@iro.umontreal.ca>,  Lars Ingebrigtsen
>  <larsi@gnus.org>,  emacs-devel@gnu.org,  Eli Zaretskii <eliz@gnu.org>,
>   Dmitry Gutov <dgutov@yandex.ru>
> Date: Wed, 15 Sep 2021 19:02:52 +0300
> 
> Why not to use proper Unicode quotation like ‘...’?

Because not every terminal can display those.

We've been through this argument already, and that is why we decided
to use '..'.

I don't really see why people keep arguing, since everyone around here
uses the quoting they like anyway, disregarding the guidelines.  I see
`..', '..', and `..` all the time.  It's anarchy anyway, so why people
insist on arguing about something they won't follow anyway?  Just let
it go.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 15.09.2021 19:30, Eli Zaretskii wrote:
> I don't really see why people keep arguing, since everyone around here
> uses the quoting they like anyway, disregarding the guidelines.

Your commit broke symbol highlighting in the comments.

That's why I spoke up originally.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: rudalics@gmx.at, monnier@iro.umontreal.ca, larsi@gnus.org,
>  emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Wed, 15 Sep 2021 21:07:06 +0300
> 
> On 15.09.2021 19:30, Eli Zaretskii wrote:
> > I don't really see why people keep arguing, since everyone around here
> > uses the quoting they like anyway, disregarding the guidelines.
> 
> Your commit broke symbol highlighting in the comments.

And `this` doesn't?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 15.09.2021 21:16, Eli Zaretskii wrote:
>> Cc:rudalics@gmx.at,monnier@iro.umontreal.ca,larsi@gnus.org,
>>   emacs-devel@gnu.org
>> From: Dmitry Gutov<dgutov@yandex.ru>
>> Date: Wed, 15 Sep 2021 21:07:06 +0300
>>
>> On 15.09.2021 19:30, Eli Zaretskii wrote:
>>> I don't really see why people keep arguing, since everyone around here
>>> uses the quoting they like anyway, disregarding the guidelines.
>> Your commit broke symbol highlighting in the comments.
> And `this` doesn't?

It also does. If I see this style of quoting in Elisp comments (which I 
have not, so far), I will correct it to `...'.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > And `this` doesn't?
> 
> It also does. If I see this style of quoting in Elisp comments (which I
> have not, so far), I will correct it to `...'.

+1.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

> > Your commit broke symbol highlighting in the comments.
>
> And `this` doesn't?

Yes, it does.  That's why I originally changed it to `this'.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Wed, 15 Sep 2021 20:40:06 +0200
> Cc: Juri Linkov <juri@linkov.net>, martin rudalics <rudalics@gmx.at>,
>  Stefan Monnier <monnier@iro.umontreal.ca>, Dmitry Gutov <dgutov@yandex.ru>,
>  Lars Ingebrigtsen <larsi@gnus.org>, Emacs developers <emacs-devel@gnu.org>
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > > Your commit broke symbol highlighting in the comments.
> >
> > And `this` doesn't?
> 
> Yes, it does.  That's why I originally changed it to `this'.

yes, everyone seems to do what they like in this respect.  So I did as
well (except that in my case I have a discussion with a conclusion to
back me up, not that it matters).

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 15.09.2021 21:52, Eli Zaretskii wrote:
>> From: Stefan Kangas<stefan@marxist.se>
>> Date: Wed, 15 Sep 2021 20:40:06 +0200
>> Cc: Juri Linkov<juri@linkov.net>, martin rudalics<rudalics@gmx.at>,
>>   Stefan Monnier<monnier@iro.umontreal.ca>, Dmitry Gutov<dgutov@yandex.ru>,
>>   Lars Ingebrigtsen<larsi@gnus.org>, Emacs developers<emacs-devel@gnu.org>
>>
>> Eli Zaretskii<eliz@gnu.org>  writes:
>>
>>>> Your commit broke symbol highlighting in the comments.
>>> And `this` doesn't?
>> Yes, it does.  That's why I originally changed it to `this'.
> yes, everyone seems to do what they like in this respect.

One is not like the other.

> So I did as
> well (except that in my case I have a discussion with a conclusion to
> back me up, not that it matters).

Then, if you don't mind, I will revert your commit.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

> Then, if you don't mind, I will revert your commit.

I think we should first try to agree on a convention (specifically for
quoting symbols in ELisp comments).  Either:

- We have no official preference (in which case the maintainer of
  a package/file can try and impose his preference for that file, but
  beyond this we should just refrain from changing other people's
  style).  I guess that would mean you shouldn't revert his commit.
  Not sure what that would mean for font-lock and symbol completion.

- We declare '...' as the official style and decide on how we deal with
  the many cases of other style used throughout the code, and someone™
  will have to write the new code for font-lock and completion.

- We declare `...' as the official style.  This matches the historical
  use, so it requires no changes, AFAIK (other than reverting Eli's
  commit, obviously).

- We declare something else as the official style.  I personally have
  grown fond of (a subset of) Markdown, so I'd vote for `...` (and
  encourage extending this to a few more conventions, e.g. so we can
  distinguish code from prose in comments), but (a subset of) Org style
  would also make a lot of sense.

This is a typical bikeshed kind of question, so there's not much point
getting people to argue over it.  The maintainers should make an
executive decision on it (but they do need to come to an agreement
among themselves ;-).


        Stefan

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> - We declare `...' as the official style.  This matches the historical
>   use, so it requires no changes, AFAIK (other than reverting Eli's
>   commit, obviously).

This, please.

`...' was _long_ the practice, and a very good one.
Simple to fontify, simple to search for, not very
confusable with anything else.  Simple, easy-peasy,
straightforward.

It was the practice in the Emacs, Elisp and Eintr
(and more) manuals.  And in doc strings.  And in
Lisp comments.  So simple.

Then came the curly-quotes decorator wave against
a perceived (by some) primitive ugliness...  And
now it seems nothing's defined - things are
apparently up for grabs.

An unfortunate change was made, despite (at least
some) expressed disapproval.  And now it's all
"bike-shedding" - a free-for-all.  Have at it.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: Eli Zaretskii <eliz@gnu.org>,  Stefan Kangas <stefan@marxist.se>,
>   juri@linkov.net,  rudalics@gmx.at,  larsi@gnus.org,  emacs-devel@gnu.org
> Date: Wed, 15 Sep 2021 16:15:59 -0400
> 
> This is a typical bikeshed kind of question, so there's not much point
> getting people to argue over it.  The maintainers should make an
> executive decision on it (but they do need to come to an agreement
> among themselves ;-).

We already had this discussion, and there was a conclusion.  We
decided to use '..' as the quoting style in plain-text documents.
I've been using it everywhere since then.  The main arguments were:

  . the ` character looks ugly on many modern displays
  . the Unicode quotes are not universally supported on text-mode terminals
  . therefore '..' has the least problems

This discussion was a kind-of tangent of the dispute triggered by the
changes in how we display `foo' quoted symbols in doc strings and
echo-area messages.

If we want '..' to be highlighted in some contexts, let's change
font-lock to support that.

And please remember that we are not talking only about comments, nor
only about ELisp files.  We are talking about all kinds of plain-text
documentation: comments, Git commit log messages, NEWS, PROBLEMS, text
files in admin/notes/, etc.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 08:08, Eli Zaretskii wrote:
>> From: Stefan Monnier <monnier@iro.umontreal.ca>
>> Cc: Eli Zaretskii <eliz@gnu.org>,  Stefan Kangas <stefan@marxist.se>,
>>    juri@linkov.net,  rudalics@gmx.at,  larsi@gnus.org,  emacs-devel@gnu.org
>> Date: Wed, 15 Sep 2021 16:15:59 -0400
>>
>> This is a typical bikeshed kind of question, so there's not much point
>> getting people to argue over it.  The maintainers should make an
>> executive decision on it (but they do need to come to an agreement
>> among themselves ;-).
> 
> We already had this discussion, and there was a conclusion.  We
> decided to use '..' as the quoting style in plain-text documents.
> I've been using it everywhere since then.  The main arguments were:
> 
>    . the ` character looks ugly on many modern displays

The "` character look ugly" discussion resulted in changes of how these 
characters are rendered in the manual, the Help buffers, and so on.

And in this manual page:

https://www.gnu.org/software/emacs/manual/html_node/elisp/Documentation-Tips.html

Note how it still recommends either using curved quotes around symbols 
in source code (which you apparently dislike; and so do I), or keep to 
the old convention of `..'.

>    . the Unicode quotes are not universally supported on text-mode terminals
>    . therefore '..' has the least problems
> 
> This discussion was a kind-of tangent of the dispute triggered by the
> changes in how we display `foo' quoted symbols in doc strings and
> echo-area messages.

That tangent was about NEWS and ChangeLogs and other plain text files.

> If we want '..' to be highlighted in some contexts, let's change
> font-lock to support that.

You can't just go around breaking things and expect others to pick up 
the pieces.

> And please remember that we are not talking only about comments, nor
> only about ELisp files.  We are talking about all kinds of plain-text
> documentation: comments, Git commit log messages, NEWS, PROBLEMS, text
> files in admin/notes/, etc.

No, we are not.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Dmitry Gutov <dgutov@yandex.ru> writes:

>> And please remember that we are not talking only about comments, nor
>> only about ELisp files.  We are talking about all kinds of plain-text
>> documentation: comments, Git commit log messages, NEWS, PROBLEMS, text
>> files in admin/notes/, etc.
>
> No, we are not.

I agree with using '...' in plain text files, like NEWS and PROBLEMS.

.el/.c files are not plain text files, they are source code files, and I
do not agree that we have to (or should) change the convention from
`...'  there.  At least not until we have fixed our tool chain to
respect '...'  in .el/.c files -- and if we decide to move to '...', we
should do so everywhere in those files: Both doc strings and comments.

Switching to anything new here seems to have marginal utility for such a
pervasive change, so I don't see much point.  I don't think the `
character is "ugly" -- that Markdown (for instance) uses it for similar
purposes (but as `...`) shows that it's a character that people are
comfortable with using in semi-structured non-plain-text files like
source code files.

(And when it comes to commit messages, I don't think we should have a
convention -- it just doesn't matter, and people should use whatever
they want to.)

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: Eli Zaretskii <eliz@gnu.org>,  Stefan Monnier
>  <monnier@iro.umontreal.ca>,  stefan@marxist.se,  juri@linkov.net,
>   rudalics@gmx.at,  emacs-devel@gnu.org
> Date: Thu, 16 Sep 2021 14:32:30 +0200
> 
> Dmitry Gutov <dgutov@yandex.ru> writes:
> 
> >> And please remember that we are not talking only about comments, nor
> >> only about ELisp files.  We are talking about all kinds of plain-text
> >> documentation: comments, Git commit log messages, NEWS, PROBLEMS, text
> >> files in admin/notes/, etc.
> >
> > No, we are not.
> 
> I agree with using '...' in plain text files, like NEWS and PROBLEMS.

If this discussion indicates something it's that "plain text" is open
to interpretation, so if we will have a mandatory convention, we'd
need to come up with an explicit list.  E.g., NEWS and PROBLEMS are by
default in Outline mode, not in Text mode, so someone could claim they
aren't "plain text".

> .el/.c files are not plain text files, they are source code files, and I
> do not agree that we have to (or should) change the convention from
> `...'  there.  At least not until we have fixed our tool chain to
> respect '...'  in .el/.c files -- and if we decide to move to '...', we
> should do so everywhere in those files: Both doc strings and comments.

Well, we already have hundreds of quotes 'like this' in both Lisp and
C files, so if highlighting that is a problem, let's work on that,
because reverting all those hundreds of quotes sounds like a lot of
code churn.  (Btw, C and Lisp are not the only programming languages
we use in Emacs sources, is the above valid for all the source files?)

> I don't think the ` character is "ugly" -- that Markdown (for
> instance) uses it for similar purposes (but as `...`) shows that
> it's a character that people are comfortable with using in
> semi-structured non-plain-text files like source code files.

We've had this argument, and I'd prefer not to have it again.
(Markdown is not really relevant because `...` does not display with
these characters.)

> (And when it comes to commit messages, I don't think we should have a
> convention -- it just doesn't matter, and people should use whatever
> they want to.)

We produce ChangeLog files from that.

And if we allow people to do what they want, then we have that
already, so I see no reason to argue about conventions no one will
follow.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 15:53, Eli Zaretskii wrote:

>> I agree with using '...' in plain text files, like NEWS and PROBLEMS.
> 
> If this discussion indicates something it's that "plain text" is open
> to interpretation, so if we will have a mandatory convention, we'd
> need to come up with an explicit list.  E.g., NEWS and PROBLEMS are by
> default in Outline mode, not in Text mode, so someone could claim they
> aren't "plain text".

They are not program code. But they can have excerpts.

>> .el/.c files are not plain text files, they are source code files, and I
>> do not agree that we have to (or should) change the convention from
>> `...'  there.  At least not until we have fixed our tool chain to
>> respect '...'  in .el/.c files -- and if we decide to move to '...', we
>> should do so everywhere in those files: Both doc strings and comments.
> 
> Well, we already have hundreds of quotes 'like this' in both Lisp and
> C files, so if highlighting that is a problem, let's work on that,
> because reverting all those hundreds of quotes sounds like a lot of
> code churn.  (Btw, C and Lisp are not the only programming languages
> we use in Emacs sources, is the above valid for all the source files?)

Are you arguing toward having multiple standards or no standard at all?

I just did a search across .el files:

   rg --type lisp ';.*?[^=]\'[a-z][a-z_-]{6,}\''

results in 199 lines, a lot of which are "plain" quotations by 
themselves, e.g.

test/lisp/so-long-tests/so-long-tests.el:
364:  ;; is 'preserved' by default (using both options).
498:         ;; will validate the 'reverted' state against this.  (Note

or

lisp/vc/pcvs.el
85:;;    add support for parsing 'modules' file ("cvs co -c")
97:;; - write 'cvs-mode-admin' to do arbitrary 'cvs admin' commands

which don't require changing. Others refer to symbols but not global 
symbols (and thus don't benefit from code completion/navigation support 
so much), so not using proper markup could be excused.

Either way, that is less than 200 examples, easy enough to undo.

For comparison,

   rg --type lisp ';.*?[^=]`[a-z][a-z_-]{6,}\''

Reports 10691 lines.

>> I don't think the ` character is "ugly" -- that Markdown (for
>> instance) uses it for similar purposes (but as `...`) shows that
>> it's a character that people are comfortable with using in
>> semi-structured non-plain-text files like source code files.
> 
> We've had this argument, and I'd prefer not to have it again.
> (Markdown is not really relevant because `...` does not display with
> these characters.)

Where doesn't it display with these characters? In the source files? Yes 
it does.

But like Stefan K. said, we could choose font-lock rules to have them 
displayed however you want. That's not the issue.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Phil Sainty]

On 2021-09-17 01:37, Dmitry Gutov wrote:
> I just did a search across .el files:
> results in 199 lines, a lot of which are "plain" quotations by 
> themselves, e.g.
> 
> test/lisp/so-long-tests/so-long-tests.el:
> 364:  ;; is 'preserved' by default (using both options).
> 498:         ;; will validate the 'reverted' state against this.  (Note

Which for me is another argument for not changing how `symbols' are
quoted, because then we can't differentiate between these two quite
distinct uses:

If I'm writing plain English and want to quote something for emphasis
I'll use 'single' quotation marks.  If I'm writing elisp comments and
want to refer to a symbol, I'll use `symbol' quoting which gives it
the appropriate face in emacs-lisp-mode.  Personally, I *want* those
two things to look different, because when I see `symbol' highlighted
in a comment it's telling me something about the code, rather than
merely providing emphasis.

As an author I also appreciate the consistency of always using the
same quoting style for symbols within the same file.  Having to switch
between (and think about) two styles within the same file adds some
extra cognitive load (however minor) that seems to me both unnecessary
and in service of an end result which is slightly worse than what we
had to begin with.


-Phil

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 17.09.2021 02:20, Phil Sainty wrote:
> Which for me is another argument for not changing how `symbols' are
> quoted, because then we can't differentiate between these two quite
> distinct uses:
> 
> If I'm writing plain English and want to quote something for emphasis
> I'll use 'single' quotation marks.  If I'm writing elisp comments and
> want to refer to a symbol, I'll use `symbol' quoting which gives it
> the appropriate face in emacs-lisp-mode.  Personally, I *want* those
> two things to look different, because when I see `symbol' highlighted
> in a comment it's telling me something about the code, rather than
> merely providing emphasis.

100%.

This is what I've referred to as difference between punctuation and 
"markup" syntax.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> Well, we already have hundreds of quotes 'like this' in both Lisp and
> C files, so if highlighting that is a problem, let's work on that,
> because reverting all those hundreds of quotes sounds like a lot of
> code churn.

Speaking of Lisp files (I've no opinion about C etc.),
it was a waste of time (it was churn) to have purged
the files of `...' in the first place.

"Let's work on that" by reverting the misguided
(IMO) switch from `...' to '...'.
 
> > I don't think the ` character is "ugly" -- that Markdown (for
> > instance) uses it for similar purposes (but as `...`) shows that
> > it's a character that people are comfortable with using in
> > semi-structured non-plain-text files like source code files.
> 
> We've had this argument, and I'd prefer not to have it again.

No need to repeat the argument - I agree.  Let's
just reverse the reversal - get back to using
`...'.

> And if we allow people to do what they want, then we have that
> already, so I see no reason to argue about conventions no one will
> follow.

People followed the `...' convention pretty well
for decades, no?  It's simple to type (vs curly),
simple to recognize/parse, simple to fontify,
simple to search for.  It's wonderful.

What's wrong with it?  Someone thinks it's ugly,
and people new to Emacs aren't used to it.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Eli Zaretskii <eliz@gnu.org> writes:

> If this discussion indicates something it's that "plain text" is open
> to interpretation, so if we will have a mandatory convention, we'd
> need to come up with an explicit list.  E.g., NEWS and PROBLEMS are by
> default in Outline mode, not in Text mode, so someone could claim they
> aren't "plain text".

I guess somebody could -- we'll have to wait and see whether anybody
does make that argument.

>> (And when it comes to commit messages, I don't think we should have a
>> convention -- it just doesn't matter, and people should use whatever
>> they want to.)
>
> We produce ChangeLog files from that.
>
> And if we allow people to do what they want, then we have that
> already, so I see no reason to argue about conventions no one will
> follow.

You were the one that brought up commit messages, and I just re-stated
my position that we don't need to have a convention there (and you seem
to be agreeing here).

That doesn't mean that I don't think we should have a convention
elsewhere, and I think it should be, basically, what we have today:

NEWS, CONTRIBUTE and other files of a similar ilk should use '...'.
.el/.c files should use `...' (when talking about Lisp symbols).

> Btw, `...' in *.c files is not highlighted.

I'd forgotten -- it'd be nice to have that fixed.

>> and if we decide to move to '...', we
>> should do so everywhere in those files: Both doc strings and comments.
>
> That is impossible, because doc strings must still use `...'.

We'd have to change the machinery to do the same thing with '...' as we
today do with `...', of course.  (But note that I'm not advocating for
this.)

> And I don't see why comments and doc strings should use the same
> convention, they are used differently.

We often move text between comments and doc strings (mostly by making
comments into doc strings), and it seems pretty awkward to insist on
using different conventions in the two different parts of the same
source file.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: dgutov@yandex.ru,  monnier@iro.umontreal.ca,  stefan@marxist.se,
>   juri@linkov.net,  rudalics@gmx.at,  emacs-devel@gnu.org
> Date: Fri, 17 Sep 2021 16:12:05 +0200
> 
> >> (And when it comes to commit messages, I don't think we should have a
> >> convention -- it just doesn't matter, and people should use whatever
> >> they want to.)
> >
> > We produce ChangeLog files from that.
> >
> > And if we allow people to do what they want, then we have that
> > already, so I see no reason to argue about conventions no one will
> > follow.
> 
> You were the one that brought up commit messages, and I just re-stated
> my position that we don't need to have a convention there (and you seem
> to be agreeing here).

No, I don't.  I think commit log messages should use '..', as any
other plain text.

> NEWS, CONTRIBUTE and other files of a similar ilk should use '...'.
> .el/.c files should use `...' (when talking about Lisp symbols).

Which other files?  I'd like this to be specific, to avoid having this
kind of dispute in the future.  There were several people who said
they want to revert to `..' everywhere.

> >> and if we decide to move to '...', we
> >> should do so everywhere in those files: Both doc strings and comments.
> >
> > That is impossible, because doc strings must still use `...'.
> 
> We'd have to change the machinery to do the same thing with '...' as we
> today do with `...', of course.  (But note that I'm not advocating for
> this.)

Then let's drop this part, there's enough noise in the thread already.

> > And I don't see why comments and doc strings should use the same
> > convention, they are used differently.
> 
> We often move text between comments and doc strings (mostly by making
> comments into doc strings), and it seems pretty awkward to insist on
> using different conventions in the two different parts of the same
> source file.

We also move text between doc strings and the manual, or vice versa,
but no one will (I hope) suggest to use `..' instead of the Texinfo
markup and have arguments in uper-case, or vice versa.  Moving text
sometimes requires to make small adjustments, and that's not a
catastrophe.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

>> We often move text between comments and doc strings (mostly by making
>> comments into doc strings), and it seems pretty awkward to insist on
>> using different conventions in the two different parts of the same
>> source file.
>
> We also move text between doc strings and the manual, or vice versa,
> but no one will (I hope) suggest to use `..' instead of the Texinfo
> markup and have arguments in uper-case, or vice versa.

That's unavoidable though.

> Moving text sometimes requires to make small adjustments, and that's
> not a catastrophe.

Having to change quotes between docstrings and comments is not a
catastrophe, but an annoyance and a fully avoidable one at that.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Fri, 17 Sep 2021 12:21:10 -0700
> Cc: emacs-devel@gnu.org, rudalics@gmx.at, monnier@iro.umontreal.ca, 
> 	dgutov@yandex.ru, juri@linkov.net
> 
> > We also move text between doc strings and the manual, or vice versa,
> > but no one will (I hope) suggest to use `..' instead of the Texinfo
> > markup and have arguments in uper-case, or vice versa.
> 
> That's unavoidable though.

No, it isn't.  People who are proficient in Texinfo have no problem
reading Texinfo markup, so we could decide to leave it intact.  We
won't, of course, because the minor annoyance of adapting the text to
its environment is nothing compared to the other aspects.

> > Moving text sometimes requires to make small adjustments, and that's
> > not a catastrophe.
> 
> Having to change quotes between docstrings and comments is not a
> catastrophe, but an annoyance and a fully avoidable one at that.

It's a minor annoyance that is well worth sustaining.  As we already
decided.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

>> > no one will (I hope) suggest to use `..' instead of the Texinfo
>> > markup and have arguments in uper-case
>>
>> That's unavoidable though.
>
> No, it isn't.

It is unavoidable to use Texinfo markup in Texinfo files.

>> Having to change quotes between docstrings and comments is not a
>> catastrophe, but an annoyance and a fully avoidable one at that.
>
> It's a minor annoyance that is well worth sustaining.

It is avoidable to use two forms of quoting in Emacs Lisp files.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Sat, 18 Sep 2021 01:17:41 -0700
> Cc: juri@linkov.net, rudalics@gmx.at, monnier@iro.umontreal.ca,
>  dgutov@yandex.ru, larsi@gnus.org, emacs-devel@gnu.org
> 
> >> Having to change quotes between docstrings and comments is not a
> >> catastrophe, but an annoyance and a fully avoidable one at that.
> >
> > It's a minor annoyance that is well worth sustaining.
> 
> It is avoidable to use two forms of quoting in Emacs Lisp files.

It would be an annoyance to remember that text in Lisp comments should
be treated differently from other text.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Eli Zaretskii <eliz@gnu.org> writes:

> We also move text between doc strings and the manual, or vice versa,
> but no one will (I hope) suggest to use `..' instead of the Texinfo
> markup and have arguments in uper-case, or vice versa.

That's reductio ad absurdum, so of course nobody has suggested that.

> Moving text sometimes requires to make small adjustments, and that's
> not a catastrophe.

And again, nobody has said it's a catastrophe, but it's an unnecessary
annoyance.

I think we've been over all the issues here several times, and my stance
is that we should continue to use `...' (as we have for decades) to
quote Lisp symbols in .el files in doc strings and in comments.  (I'm
not particularly interested in all of the other side issues that have
come up in this thread.)

If you don't agree, we'll just have to agree to disagree.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: dgutov@yandex.ru,  monnier@iro.umontreal.ca,  stefan@marxist.se,
>   juri@linkov.net,  rudalics@gmx.at,  emacs-devel@gnu.org
> Date: Sat, 18 Sep 2021 15:30:58 +0200
> 
> I think we've been over all the issues here several times, and my stance
> is that we should continue to use `...' (as we have for decades) to
> quote Lisp symbols in .el files in doc strings and in comments.  (I'm
> not particularly interested in all of the other side issues that have
> come up in this thread.)
> 
> If you don't agree, we'll just have to agree to disagree.

And I'm not particularly interested in solving just the Lisp comments
issue.

So I guess the anarchy will continue.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > we should continue to use `...' (as we have for decades) to
> > quote Lisp symbols in .el files in doc strings and in comments...
> > If you don't agree, we'll just have to agree to disagree.
> 
> And I'm not particularly interested in solving
> just the Lisp comments issue.
> 
> So I guess the anarchy will continue.

You're the maintainer.  But is it perhaps
just an anarchy of one, or few?

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Drew Adams <drew.adams@oracle.com>
> CC: "stefan@marxist.se" <stefan@marxist.se>,
>         "emacs-devel@gnu.org"
> 	<emacs-devel@gnu.org>,
>         "rudalics@gmx.at" <rudalics@gmx.at>,
>         "monnier@iro.umontreal.ca" <monnier@iro.umontreal.ca>,
>         "dgutov@yandex.ru"
> 	<dgutov@yandex.ru>,
>         "juri@linkov.net" <juri@linkov.net>
> Date: Sat, 18 Sep 2021 16:03:39 +0000
> Accept-Language: en-US
> 
> > So I guess the anarchy will continue.
> 
> You're the maintainer.

That's not a magic wand.

> But is it perhaps just an anarchy of one, or few?

Sadly, no.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > > So I guess the anarchy will continue.
> > You're the maintainer.
> That's not a magic wand.

And yet, so far...

> > But is it perhaps just an anarchy of one, or few?
> Sadly, no.

Who's with you on this one, in the current
discussion?

I don't see multitudes proclaiming support
for changing `...' in Lisp comments (or
for keeping the change to curly quotes in
help and manuals).

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Drew Adams <drew.adams@oracle.com> writes:

> I don't see multitudes proclaiming support
> for changing `...' in Lisp comments (or
> for keeping the change to curly quotes in
> help and manuals).

Reverting how curly quotes are displayed in help and manuals has never
been on the table.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > I don't see multitudes proclaiming support
> > for changing `...' in Lisp comments (or
> > for keeping the change to curly quotes in
> > help and manuals).
> 
> Reverting how curly quotes are displayed in help
> and manuals has never been on the table.

It's not about "how curly quotes are displayed".

It's about _using_ curly quotes as delimiters
in help and manuals, where we used to use the
simpler, more recognizable, easier to search
for, and easier to type `...'.

Such reverting has been raised/proposed in the
current discussion, and not only by me.

Whether that's off the table is your (& Eli's)
decision.  But opinions in favor have been
expressed.  Does "off the table" mean it can't
even be spoken?  Maybe the opinions expressed
won't go anywhere, but still, they can be heard.

The same reasons support that reverting as
those given for not changing `...' in Lisp
comments.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Date: Thu, 16 Sep 2021 14:32:30 +0200
> Cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at,
>  Stefan Monnier <monnier@iro.umontreal.ca>, Eli Zaretskii <eliz@gnu.org>,
>  juri@linkov.net
> 
> .el/.c files are not plain text files, they are source code files, and I
> do not agree that we have to (or should) change the convention from
> `...'  there.  At least not until we have fixed our tool chain to
> respect '...'  in .el/.c files

Btw, `...' in *.c files is not highlighted.

> and if we decide to move to '...', we
> should do so everywhere in those files: Both doc strings and comments.

That is impossible, because doc strings must still use `...'.  And I
don't see why comments and doc strings should use the same convention,
they are used differently.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> doc strings must still use `...'.  And I don't see
> why comments and doc strings should use the same
> convention, they are used differently.

How are they "used differently"?  For what reason
should they NOT use the same convention?

Occam tells us not to multiply things needlessly.
What's the need here for a convention difference
between doc strings and comments?  For decades we
used `...' for both, no?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Date: Thu, 16 Sep 2021 14:32:30 +0200
> Cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at,
>  Stefan Monnier <monnier@iro.umontreal.ca>, Eli Zaretskii <eliz@gnu.org>,
>  juri@linkov.net
> 
> .el/.c files are not plain text files, they are source code files, and I
> do not agree that we have to (or should) change the convention from
> `...'  there.  At least not until we have fixed our tool chain to
> respect '...'  in .el/.c files

I've now made a simple change which causes symbols quoted 'like-this'
be highlighted in Lisp files.  OK?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 17:15, Eli Zaretskii wrote:
> I've now made a simple change which causes symbols quoted 'like-this'
> be highlighted in Lisp files.  OK?

That leaves code completion.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at,
>  monnier@iro.umontreal.ca, juri@linkov.net
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 16 Sep 2021 17:18:48 +0300
> 
> On 16.09.2021 17:15, Eli Zaretskii wrote:
> > I've now made a simple change which causes symbols quoted 'like-this'
> > be highlighted in Lisp files.  OK?
> 
> That leaves code completion.

What do you mean by that?  I didn't change anything in the completion
area.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > do not agree that we have to (or should) change the convention from
> > `...'  there.  At least not until we have fixed our tool chain to
> > respect '...'  in .el/.c files
> 
> I've now made a simple change which causes symbols quoted 'like-this'
> be highlighted in Lisp files.  OK?

Not by me, FWIW.  

It's not just about symbols.  We mention key bindings
in doc and comments, and those can contain space chars.
Likewise, we sometimes have other quoted char sequences
that contain space chars.

It's true that we haven't, so far, fontified things
like `C-x 4 f' in comments.  But we do fontified them
in Info and doc strings (albeit now using curly quotes).

E.g. in (elisp) `Disassembly' there's this: 'dup varset',
which is fontified, even though it contains a space char.
And in (elisp) `Keymap Basics' there's this: `C-f C-n'
which is fontified.  It would be good to do the same in
Lisp comments, IMO.

If you're really looking for something that would really
improve things - and not only for a beauty contest (but
yes, more beautiful too), then let's work on doing that.

Bring back `...', and fontify the delimited text even
when it's not a "symbol".

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 712 bytes --]


>
> I don't think the ` character is "ugly" -- that Markdown (for instance) 
> uses it for similar purposes (but as `...`) shows that it's a character 
> that people are comfortable with using in semi-structured non-plain-text 
> files like source code files.
>

FWIW, I too do not see why ` would be "ugly".  I attach screenshots of 
what that character looks like in ten popular programming fonts; each 
shows the nine (!) possible conventions, together with the Unicode quotes. 
What would perhaps be welcome is a user option to ask Emacs to display 
apostrophes as acutes in `foobar' (and possibly another option to ask 
Emacs to use `´ instead of ‘’ in messages, errors and help buffers).

[-- Attachment #2: Type: image/png, Size: 9751 bytes --]

[-- Attachment #3: Type: image/png, Size: 9341 bytes --]

[-- Attachment #4: Type: image/png, Size: 10528 bytes --]

[-- Attachment #5: Type: image/png, Size: 10849 bytes --]

[-- Attachment #6: Type: image/png, Size: 9619 bytes --]

[-- Attachment #7: Type: image/png, Size: 11524 bytes --]

[-- Attachment #8: Type: image/png, Size: 10019 bytes --]

[-- Attachment #9: Type: image/png, Size: 10218 bytes --]

[-- Attachment #10: Type: image/png, Size: 10276 bytes --]

[-- Attachment #11: Type: image/png, Size: 10021 bytes --]

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Gregory Heytings <gregory@heytings.org> writes:

> FWIW, I too do not see why ` would be "ugly".  I attach screenshots of
> what that character looks like in ten popular programming fonts; each
> shows the nine (!) possible conventions, together with the Unicode quotes.

This shows that Unicode quotes consistently look good, while the others
look antiquated at best.  The single apostrophes on both sides ('like
this') is also okay, but still worse from a typographic standpoint.

IMO, we should automatically display in comments and docstrings whatever
quote character the user prefers according to `text-quoting-style', and
automatically insert the correct quote character when typing "'".

At least for docstrings and messages, what would be even better than any
markers is to display prose in a variable font and code (including
symbols, of course) in a monospace font, with no markers.  That is the
style used elsewhere, and it is clear and avoids visual distractions.

> What would perhaps be welcome is a user option to ask Emacs to display
> apostrophes as acutes in `foobar' (and possibly another option to ask
> Emacs to use `´ instead of ‘’ in messages, errors and help buffers).

Are you looking for `text-quoting-style'?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Fri, 17 Sep 2021 03:26:38 -0700
> Cc: Dmitry Gutov <dgutov@yandex.ru>, emacs-devel@gnu.org, rudalics@gmx.at, 
> 	Stefan Monnier <monnier@iro.umontreal.ca>, Eli Zaretskii <eliz@gnu.org>, juri@linkov.net
> 
> Gregory Heytings <gregory@heytings.org> writes:
> 
> > FWIW, I too do not see why ` would be "ugly".  I attach screenshots of
> > what that character looks like in ten popular programming fonts; each
> > shows the nine (!) possible conventions, together with the Unicode quotes.
> 
> This shows that Unicode quotes consistently look good, while the others
> look antiquated at best.  The single apostrophes on both sides ('like
> this') is also okay, but still worse from a typographic standpoint.

I think that was the conclusion back when this was discussed, and we
decided to use the "second best" due to the issue with Unicode support
in text-only terminals.

> IMO, we should automatically display in comments and docstrings whatever
> quote character the user prefers according to `text-quoting-style', and
> automatically insert the correct quote character when typing "'".

Patches welcome to make that happen.  One idea for implementation
would be to use the display-table feature.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

>> IMO, we should automatically display in comments and docstrings whatever
>> quote character the user prefers according to `text-quoting-style', and
>> automatically insert the correct quote character when typing "'".
>
> Patches welcome to make that happen.  One idea for implementation
> would be to use the display-table feature.

I can't see any way to limit `buffer-display-table' to docstrings.
Am I missing something?

Here's what I tried:

(define-minor-mode emacs-lisp-fancy-docstring-mode
  "Minor mode for fancy display of Emacs Lisp `docstrings'."
  :init-value nil
  (if emacs-lisp-fancy-docstring-mode
      (let ((table (or (copy-sequence standard-display-table)
                       (make-display-table))))
        (aset table ?` `[,(string-to-char (substitute-command-keys "`"))])
        (aset table ?' `[,(string-to-char (substitute-command-keys "'"))])
        (setq buffer-display-table table))
    (setq buffer-display-table nil)))

It does the job in the sense that it correctly displays these quotes in
docstrings, but it also does it in code, which makes things hard to
read.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Fri, 17 Sep 2021 15:46:25 -0700
> Cc: emacs-devel@gnu.org, rudalics@gmx.at, gregory@heytings.org, 
> 	monnier@iro.umontreal.ca, dgutov@yandex.ru, larsi@gnus.org, juri@linkov.net
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Patches welcome to make that happen.  One idea for implementation
> > would be to use the display-table feature.
> 
> I can't see any way to limit `buffer-display-table' to docstrings.

You can't.

> (define-minor-mode emacs-lisp-fancy-docstring-mode
>   "Minor mode for fancy display of Emacs Lisp `docstrings'."
>   :init-value nil
>   (if emacs-lisp-fancy-docstring-mode
>       (let ((table (or (copy-sequence standard-display-table)
>                        (make-display-table))))
>         (aset table ?` `[,(string-to-char (substitute-command-keys "`"))])
>         (aset table ?' `[,(string-to-char (substitute-command-keys "'"))])
>         (setq buffer-display-table table))
>     (setq buffer-display-table nil)))
> 
> It does the job in the sense that it correctly displays these quotes in
> docstrings, but it also does it in code, which makes things hard to
> read.

If it makes things harder to read, then my idea won't fly.  I thought
that since ‘ and ’ look so very similar to ` and ', the effect on
readability of code will be negligible.

If this idea is a non-starter, and we want to do this only in comments
and strings, I guess the only reasonable alternative is to do it as
part of font-lock?  Which means people who turn off font-lock will not
be able to have the feature.

Or maybe we could introduce a 'display-table' text property, which
would then make it possible to do this only for some parts of the
buffer text.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Andreas Schwab]

On Sep 18 2021, Eli Zaretskii wrote:

> If it makes things harder to read, then my idea won't fly.  I thought
> that since ‘ and ’ look so very similar to ` and ',

Not at all.  Both ‘ and ’ are slanted to the right, making it impossible
to distinguish backquote from quote.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: Stefan Kangas <stefan@marxist.se>,  emacs-devel@gnu.org,
>   rudalics@gmx.at,  gregory@heytings.org,  monnier@iro.umontreal.ca,
>   dgutov@yandex.ru,  larsi@gnus.org,  juri@linkov.net
> Date: Sat, 18 Sep 2021 09:32:52 +0200
> 
> On Sep 18 2021, Eli Zaretskii wrote:
> 
> > If it makes things harder to read, then my idea won't fly.  I thought
> > that since ‘ and ’ look so very similar to ` and ',
> 
> Not at all.  Both ‘ and ’ are slanted to the right, making it impossible
> to distinguish backquote from quote.

What kind of font slants ‘ to the right?  Why is that a reasonable
rendering?  And how do you distinguish between ‘ and ’ with that font?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Adam Sjøgren]

[-- Attachment #1: Type: text/plain, Size: 149 bytes --]

Eli writes:

> What kind of font slants ‘ to the right?

    ftcrhb:-PfEd-DejaVu Sans Mono-normal-normal-normal-*-30-*-*-*-m-0-iso10646-1 (#x723)


[-- Attachment #2: DejaVu Sans Mono screenshot --]
[-- Type: image/png, Size: 4844 bytes --]

[-- Attachment #3: Type: text/plain, Size: 217 bytes --]


> Why is that a reasonable rendering?

I guess that's for a type designer to answer.

> And how do you distinguish between ‘ and ’ with that font?

It looks like it's whether the top or bottom is the thin part:


[-- Attachment #4: DejaVu Sans Mono screenshot 2 --]
[-- Type: image/png, Size: 3919 bytes --]

[-- Attachment #5: Type: text/plain, Size: 180 bytes --]



  Best regards,

    Adam

-- 
 "I'm a dessertarian."                                      Adam Sjøgren
                                                       asjo@koldfront.dk

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 759 bytes --]


>
> What kind of font slants ‘ to the right?
>

Many (but not all) fonts.

>
> Why is that a reasonable rendering?
>

Because the quotation mark is a transformed comma.  The right quotation 
mark is a raised comma, and the left quotation mark is an raised comma 
rotated by 180 degrees.  In many (but not all) fonts, the shape of a comma 
is more or less that of a triangle with an obtuse angle at the top left, 
that is, slanted to the right.  When you rotate such a shape by 180 
degrees, it is again slanted to the right.

>
> And how do you distinguish between ‘ and ’ with that font?
>

Because the thick part (aka the dot) of ‘ is at the bottom of the 
character, and the thick part of ’ is at the top of the character.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Yuri Khan]

On Sat, 18 Sept 2021 at 15:27, Eli Zaretskii <eliz@gnu.org> wrote:

> > Not at all.  Both ‘ and ’ are slanted to the right, making it impossible
> > to distinguish backquote from quote.

> What kind of font slants ‘ to the right?

Pretty much all of them. Liberation Mono, to name just one.

There are fonts that render ‘ as a mirrored ’, but they are a minority.

> Why is that a reasonable rendering?

Because that’s how they were traditionally rendered with printing
presses, and handwritten before that.

> And how do you distinguish between ‘ and ’ with that font?

In many fonts, the opening quote looks like a comma turned 180° (tail
up and to the right) and raised to the top of the cell. The closing
quote looks like a comma (tail down and to the left) just raised. With
serif fonts, this is like small 6 and 9 shapes; with sans-serif fonts,
the tail is thinner than the start of stroke.

For me, the difference is quite visible even on a non-HiDPI display; I
disagree that they are impossible to distinguish.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Phil Sainty]

On 2021-09-18 20:48, Yuri Khan wrote:
> On Sat, 18 Sept 2021 at 15:27, Eli Zaretskii <eliz@gnu.org> wrote:
>> And how do you distinguish between ‘ and ’ with that font?
> 
> For me, the difference is quite visible even on a non-HiDPI display;
> I disagree that they are impossible to distinguish.

That's going to vary dramatically from person to person depending on
font, screen, eyesight, etc.  I'm reading this now in such a font, and
I notice the difference between the two characters only if they are
near to one another and I'm paying close attention to them.

At a glance I find them incredibly difficult to distinguish (if these
characters were used for lisp quote and backquote, for instance, it
would be absolutely awful).


-Phil

Re: master ff4de1b: Fix quoting style in Lisp comments

[Yuri Khan]

On Sat, 18 Sept 2021 at 18:50, Phil Sainty <psainty@orcon.net.nz> wrote:

> > For me, the difference is quite visible even on a non-HiDPI display;
> > I disagree that they are impossible to distinguish.
>
> At a glance I find them incredibly difficult to distinguish (if these
> characters were used for lisp quote and backquote, for instance, it
> would be absolutely awful).

I agree ‘ and ’ should not be used for Lisp quote and backquote. In
Lisp, those aren’t even paired syntax delimiters the way form
parentheses, vector brackets, or string literal double quotes are.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> I agree ‘ and ’ should not be used for Lisp quote and backquote.

Right.

> In Lisp, those aren’t even paired syntax delimiters
> the way form parentheses, vector brackets, or string
> literal double quotes are.

Right ... so far (knock on wood).

If they were used in Elisp comments (‘...’) they'd be
paired delimiters.  Thank goodness we haven't gone
there (yet).  But it may take only another Cosmetic
Crusader to launch that crusade...  And then another
ship-sailing decision in favor of that beautification
and "modernization".

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> At a glance I find them incredibly difficult to distinguish (if these
> characters were used for lisp quote and backquote, for instance, it
> would be absolutely awful).

Exactly.  Their use in natural-language text
isn't problematic, in general.  Their use in
contexts where their meanings are important
is problematic.  For example use in regexps:

(defvar info-quotation-regexp
  (concat
   "\"\\(?:[^\"\\]\\|\\\\\\(?:.\\|\n\\)\\)*\"\\|" ; "..."
   "`\\(?:[^']\\|\\\\\\(?:.\\|\n\\)\\)+'\\|"      ; `...'
   "‘\\(?:[^’]\\|\\\\\\(?:.\\|\n\\)\\)+’\\|"      ; ‘...’
   "“\\(?:[^”]\\|\\\\\\(?:.\\|\n\\)\\)+”"         ; “...”
   )
  "Regexp to match `...', ‘...’, “...”, \"...\".
If ... contains an end char then that char must be backslashed.")

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> For me, the difference is quite visible even on a non-HiDPI display; I
> disagree that they are impossible to distinguish.

It's not that they're _impossible_ to distinguish.
It's that they can be _difficult_ to distinguish.

Much of the time, i.e., in may contexts, it's not
really important to be able to distinguish them.

But in text where the difference matters, including
code (whether embedded text, comments, or on its
own), ease in recognition is important.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > Both ‘ and ’ are slanted to the right, making it
> > impossible to distinguish backquote from quote.
> 
> What kind of font slants ‘ to the right?  Why is that a reasonable
> rendering?  And how do you distinguish between ‘ and ’ with that font?

With some difficulty.  I zoom in.  And it's not
just some odd font.  The default font for plain
text in my mail client (Outlook) makes them
almost indistinguishable, for example.

. and , are also difficult to distinguish, for
the same reason.  But their usage in natural
language text (and in code often) typically
makes it unnecessary to pay close attention to
them visually.

You might say the same is true for ‘...’ in
Emacs, and that's no doubt true.  The
"recognition" difficulty introduced by this
step backward is less important than the other
problems I (and others) mentioned: search,
fontifying, and typing.

Even so, difficult recognition of ‘ vs ’ _is_
a problem.  It may not be bad enough that I
have to zoom in for most uses of ‘...’, but
when checking those chars in regexps etc. I
do zoom in (it's easy to zoom in, of course).

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> Both ‘ and ’ are slanted to the right, making
> it impossible to distinguish backquote from quote.

It no doubt depends on the font used, but I agree
with this.

With fonts I use, those chars are much more
difficult to distinguish than ` and ', by a
_long_ shot.  If I really need to distinguish
them then I find myself zooming in on the text.

This is part of what I meant by `...' providing
easier "recognition".

Beauty's in the eye of the beholder, and so are
convenience and usability.  The use of curly
quotes in the doc and help was a real step
backward, IMO. I sure hope we don't extend that
purported "modernization" to comments in Elisp.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

> If this idea is a non-starter, and we want to do this only in comments
> and strings, I guess the only reasonable alternative is to do it as
> part of font-lock?  Which means people who turn off font-lock will not
> be able to have the feature.

Yes, I guess font-lock is unavoidable here.

BTW, as a minor extension to what I did above, I guess we could do
something like this if we want Unicode quotes in files like NEWS:

(define-minor-mode fancy-quotes-mode "Fancy quotes."
  :init-value nil
  (if fancy-quotes-mode
      (let ((table (or (copy-sequence standard-display-table)
                       (make-display-table))))
        (aset table ?` `[,(string-to-char (substitute-command-keys "`"))])
        (aset table ?' `[,(string-to-char (substitute-command-keys "'"))])
        (aset table ?\" [?”])
        (setq buffer-display-table table))
    (setq buffer-display-table nil)))

(add-hook 'outline-mode-hook
          (lambda ()
            (if (equal (file-name-base (buffer-file-name)) "NEWS")
                (fancy-quotes-mode))))

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Sat, 18 Sep 2021 01:17:35 -0700
> Cc: emacs-devel@gnu.org, rudalics@gmx.at, gregory@heytings.org, 
> 	monnier@iro.umontreal.ca, dgutov@yandex.ru, larsi@gnus.org, juri@linkov.net
> 
> BTW, as a minor extension to what I did above, I guess we could do
> something like this if we want Unicode quotes in files like NEWS:

We don't want Unicode quotes in NEWS, because NEWS (and other text
files we have, but NEWS, PROBLEMS, and DEBUG in particular) needs to
be readable on any text-mode terminal.  It's important to let users
read those without problems.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

>> BTW, as a minor extension to what I did above, I guess we could do
>> something like this if we want Unicode quotes in files like NEWS:
>
> We don't want Unicode quotes in NEWS, because NEWS (and other text
> files we have, but NEWS, PROBLEMS, and DEBUG in particular) needs to
> be readable on any text-mode terminal.  It's important to let users
> read those without problems.

So you would make it conditional on `display-graphic-p' or something to
that effect.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Sat, 18 Sep 2021 02:18:49 -0700
> Cc: emacs-devel@gnu.org, rudalics@gmx.at, gregory@heytings.org, 
> 	monnier@iro.umontreal.ca, dgutov@yandex.ru, larsi@gnus.org, juri@linkov.net
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> >> BTW, as a minor extension to what I did above, I guess we could do
> >> something like this if we want Unicode quotes in files like NEWS:
> >
> > We don't want Unicode quotes in NEWS, because NEWS (and other text
> > files we have, but NEWS, PROBLEMS, and DEBUG in particular) needs to
> > be readable on any text-mode terminal.  It's important to let users
> > read those without problems.
> 
> So you would make it conditional on `display-graphic-p' or something to
> that effect.

No, because we want these files to be readable by "other programs".
They are useful for those who don't (yet) have Emacs.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> We don't want Unicode quotes in NEWS, because NEWS (and other text
> files we have, but NEWS, PROBLEMS, and DEBUG in particular) needs to
> be readable on any text-mode terminal.  It's important to let users
> read those without problems.

Doesn't the same need apply to help & doc (manuals)?

"important to read without problems" is one of the
points raised against the change to curly quotes -
the "recognition" problem.

Add to that problem the difficulty of typing those
chars and searching for them...

What's good for the NEWS, PROBLEMS, and DEBUG
goose is also good for the help and doc gander.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Yuri Khan]

On Sat, 18 Sept 2021 at 15:18, Stefan Kangas <stefan@marxist.se> wrote:

> BTW, as a minor extension to what I did above, I guess we could do
> something like this if we want Unicode quotes in files like NEWS:
>
>         (aset table ?\" [?”])

If I’m not mistaken, this turns every double quote into the closing
double quote?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Yuri Khan <yuri.v.khan@gmail.com> writes:

> On Sat, 18 Sept 2021 at 15:18, Stefan Kangas <stefan@marxist.se> wrote:
>
>> BTW, as a minor extension to what I did above, I guess we could do
>> something like this if we want Unicode quotes in files like NEWS:
>>
>>         (aset table ?\" [?”])
>
> If I’m not mistaken, this turns every double quote into the closing
> double quote?

Yes, that's a limitation.

I don't see a way to do better if we want to use display tables.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Yuri Khan]

On Sat, 18 Sept 2021 at 17:05, Stefan Kangas <stefan@marxist.se> wrote:

> >> BTW, as a minor extension to what I did above, I guess we could do
> >> something like this if we want Unicode quotes in files like NEWS:
> >>
> >>         (aset table ?\" [?”])
> >
> > If I’m not mistaken, this turns every double quote into the closing
> > double quote?
>
> Yes, that's a limitation.
>
> I don't see a way to do better if we want to use display tables.

Then it would be a step from ugly simple to ugly complicated, so a net negative.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Yuri Khan <yuri.v.khan@gmail.com> writes:

> Then it would be a step from ugly simple to ugly complicated, so a net negative.

In proper Swedish we actually use right quotes on both sides of a
quote, so I don't personally find this style particularly ugly.  :-)

But note that I'm not suggesting this for inclusion.  I'm just
pointing out what can be done with display tables, as the feature was
suggested.  We would probably need to look at font-lock to find a more
complete solution.

(But I did put something much like what I proposed on my local
`text-mode-hook' already, and I find it to be an improvement.)

Re: master ff4de1b: Fix quoting style in Lisp comments

[Juri Linkov]

> But note that I'm not suggesting this for inclusion.  I'm just
> pointing out what can be done with display tables, as the feature was
> suggested.  We would probably need to look at font-lock to find a more
> complete solution.

Or ‘prettify-symbols-mode’.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Juri Linkov <juri@linkov.net>
> Cc: Yuri Khan <yuri.v.khan@gmail.com>,  Emacs developers
>  <emacs-devel@gnu.org>,  martin rudalics <rudalics@gmx.at>,  Gregory
>  Heytings <gregory@heytings.org>,  Stefan Monnier
>  <monnier@iro.umontreal.ca>,  Dmitry Gutov <dgutov@yandex.ru>,  Eli
>  Zaretskii <eliz@gnu.org>,  Lars Magne Ingebrigtsen <larsi@gnus.org>
> Date: Sat, 18 Sep 2021 21:40:27 +0300
> 
> > But note that I'm not suggesting this for inclusion.  I'm just
> > pointing out what can be done with display tables, as the feature was
> > suggested.  We would probably need to look at font-lock to find a more
> > complete solution.
> 
> Or ‘prettify-symbols-mode’.

Over my dead body.  prettify-symbols-mode uses static compositions,
which don't support bidirectional display, and so it's a non-starter
for human-readable text, such as comments and strings.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

> Over my dead body.  prettify-symbols-mode uses static compositions,
> which don't support bidirectional display, and so it's a non-starter
> for human-readable text, such as comments and strings.

I didn't study any of these features yet, so please bare with me as I'm
trying to get my bearings.

You have said before: "So prettify-symbols-mode is semi-okay, as it is
usually turned on in PL modes, where everything is left-to-right."[1]

Does that apply here as well, given that we are discussing files
containing ELisp source code?

And what is the alternative to static compositions?  The `display' text
property?

Footnotes:
[1]  https://lists.gnu.org/archive/html/emacs-devel/2021-03/msg00227.html

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> please bare with me

I really hope you meant "bear"... ;-)

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Drew Adams <drew.adams@oracle.com>
> CC: "emacs-devel@gnu.org" <emacs-devel@gnu.org>,
>         "rudalics@gmx.at"
> 	<rudalics@gmx.at>,
>         "gregory@heytings.org" <gregory@heytings.org>,
>         "monnier@iro.umontreal.ca" <monnier@iro.umontreal.ca>,
>         "dgutov@yandex.ru"
> 	<dgutov@yandex.ru>,
>         "larsi@gnus.org" <larsi@gnus.org>,
>         "yuri.v.khan@gmail.com" <yuri.v.khan@gmail.com>
> Date: Sat, 18 Sep 2021 23:22:35 +0000
> Accept-Language: en-US
> 
> > please bare with me
> 
> I really hope you meant "bear"... ;-)

Would you like Stefan to write in his first language instead?  I'm
sure he will make much fewer embarrassing mistakes if he does.

Sarcasm aside, please refrain from such insulting remarks here.  And
no, the smiley doesn't cut it.

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

>> > please bare with me
>>
>> I really hope you meant "bear"... ;-)

You are correct!  Don’t walk on glass with your bear feet.

> Would you like Stefan to write in his first language instead?  I'm
> sure he will make much fewer embarrassing mistakes if he does.

I personally found the above more funny than embarrassing.  It's really
just a typo, it happens to natives as well.

But everyone's different here, so that's something to keep in mind.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > > please bare with me
> > I really hope you meant "bear"... ;-)
> 
> Would you like Stefan to write in his first
> language instead?  I'm sure he will make much
> fewer embarrassing mistakes if he does.
> 
> Sarcasm aside, please refrain from such insulting
> remarks here.  And no, the smiley doesn't cut it.

There's nothing for anyone to be embarrassed
about.  We all type such typos, all the time.
This one is fun and funny.  And no, there was
no insult to anyone.

Lighten up please, Eli.  Have fun.  Learn
about puns and humor and pitfalls, and maybe
practice smiling or even laughing a bit.
Take a toke or a sip, if that helps.

You aggressively try to pick a fight at the
slightest pretext, with no provocation.  Stop
being so defensive - no one benefits by that
(including yourself).

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Drew Adams <drew.adams@oracle.com>
> CC: "stefan@marxist.se" <stefan@marxist.se>,
>         "juri@linkov.net"
> 	<juri@linkov.net>,
>         "emacs-devel@gnu.org" <emacs-devel@gnu.org>,
>         "rudalics@gmx.at" <rudalics@gmx.at>,
>         "gregory@heytings.org"
> 	<gregory@heytings.org>,
>         "monnier@iro.umontreal.ca"
> 	<monnier@iro.umontreal.ca>,
>         "dgutov@yandex.ru" <dgutov@yandex.ru>,
>         "larsi@gnus.org" <larsi@gnus.org>,
>         "yuri.v.khan@gmail.com"
> 	<yuri.v.khan@gmail.com>
> Date: Sun, 19 Sep 2021 17:30:11 +0000
> 
> Lighten up please, Eli.  Have fun.  Learn
> about puns and humor and pitfalls, and maybe
> practice smiling or even laughing a bit.
> Take a toke or a sip, if that helps.

Thank you for mocking me.  Regardless, please stop remarking people
about their typos.  It is inappropriate in this forum.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > Lighten up please, Eli.  Have fun.  Learn
> > about puns and humor and pitfalls, and maybe
> > practice smiling or even laughing a bit.
> > Take a toke or a sip, if that helps.
> 
> Thank you for mocking me.  

Proof of your defensiveness.

I wasn't mocking you.  Believe it or not,
I was trying to help you (and the overall
social environment/interaction).

I'd gladly share a beer with you and have
a relaxed conversation.  Bear-fighting
instead of beer-sharing is not my idea of
fun or comradery.  Life is too short for
such things.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Jean Louis]

On September 19, 2021 5:54:57 PM UTC, Drew Adams <drew.adams@oracle.com> wrote:
>> > Lighten up please, Eli.  Have fun.  Learn
>> > about puns and humor and pitfalls, and maybe
>> > practice smiling or even laughing a bit.
>> > Take a toke or a sip, if that helps.
>> 
>> Thank you for mocking me.  
>
>Proof of your defensiveness.
>
>I wasn't mocking you.  Believe it or not,
>I was trying to help you (and the overall
>social environment/interaction).
>
>I'd gladly share a beer with you and have
>a relaxed conversation.  Bear-fighting
>instead of beer-sharing is not my idea of
>fun or comradery.  Life is too short for
>such things.

Well said, you express yourself well. Eli is easily offended.

Keep clarifying, though it may be hard for such mind to understand.


Jean

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[tomas]

[-- Attachment #1: Type: text/plain, Size: 745 bytes --]

On Tue, Sep 21, 2021 at 02:38:54PM +0000, Jean Louis wrote:

[...]

> Well said, you express yourself well. Eli is easily offended.

now this is totally unnecessary.

> Keep clarifying, though it may be hard for such mind to understand.

and this borderline insulting.

FWIW, I side with Eli on that. Poking jokes at someone based on a
typo or on a language error may be funny and then it may hurt.

What it does depends on the receiving side, and the sending side
has little control over that. So it seems not a wise behaviour
before an audience as diverse as this list is.

Why take that risk? By all means, do that at home, in your circle
of friends, where you know all your interlocutors and know you
aren't going to hurt them.

Cheers
 - t

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[tomas]

[-- Attachment #1: Type: text/plain, Size: 366 bytes --]

On Tue, Sep 21, 2021 at 07:04:37PM +0200, tomas@tuxteam.de wrote:

[...]

> Why take that risk? By all means, do that at home, in your circle
> of friends [...]

And oh, what I forgot to add: the non-native speaker lurking in this
list may even grow more wary of posting anything, for fear of being
targeted by a joke, too.

I don't think we want that.

Cheers
 - t

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > FWIW, I side with Eli on that. Poking jokes at someone based on a
  > typo or on a language error may be funny and then it may hurt.

I support what you have said.

See also https://gnu.org/philosophy/kind-communication.html.

-- 
Dr Richard Stallman (https://stallman.org)
Chief GNUisance of the GNU Project (https://gnu.org)
Founder, Free Software Foundation (https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Kangas <stefan@marxist.se>
> Date: Sat, 18 Sep 2021 12:34:02 -0700
> Cc: yuri.v.khan@gmail.com, emacs-devel@gnu.org, rudalics@gmx.at, 
> 	gregory@heytings.org, monnier@iro.umontreal.ca, dgutov@yandex.ru, 
> 	larsi@gnus.org
> 
> You have said before: "So prettify-symbols-mode is semi-okay, as it is
> usually turned on in PL modes, where everything is left-to-right."[1]

For rendering _code_, i.e, symbols, yes.  But here we are talking
about comments and strings, which are a different matter altogether:
they are human-readable text, and we should not prevent people from
writing that text using bidirectional scripts.  By contrast, the
probability of someone calling a program symbol by a name with R2L
characters is ... remote at best.

> And what is the alternative to static compositions?  The `display' text
> property?

If you want to display one character as another, either 'display' text
properties (with string values) or overlay strings.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 1187 bytes --]


>> FWIW, I too do not see why ` would be "ugly".  I attach screenshots of 
>> what that character looks like in ten popular programming fonts; each 
>> shows the nine (!) possible conventions, together with the Unicode 
>> quotes.
>
> This shows that Unicode quotes consistently look good, while the others 
> look antiquated at best.  The single apostrophes on both sides ('like 
> this') is also okay, but still worse from a typographic standpoint.
>

Others might have a different opinion, depending on the font they use.

>> What would perhaps be welcome is a user option to ask Emacs to display 
>> apostrophes as acutes in `foobar' (and possibly another option to ask 
>> Emacs to use `´ instead of ‘’ in messages, errors and help buffers).
>
> Are you looking for `text-quoting-style'?
>

Not really.  Part of what I suggest would be solved with another option to 
text-quoting-style.  What we have no is ‘this’, 'this' and `this'.  What 
we could add is `this´.  And the other part of what I suggest is to have, 
say, a buffer-quoting-style option, which would have the same effect in 
regular buffers as the text-quoting-style in help buffers.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Date: Fri, 17 Sep 2021 10:55:01 +0000
> From: Gregory Heytings <gregory@heytings.org>
> cc: Lars Ingebrigtsen <larsi@gnus.org>, Dmitry Gutov <dgutov@yandex.ru>, 
>     emacs-devel@gnu.org, rudalics@gmx.at, 
>     Stefan Monnier <monnier@iro.umontreal.ca>, Eli Zaretskii <eliz@gnu.org>, 
>     juri@linkov.net
> 
> What we have no is ‘this’, 'this' and `this'.  What we could add is
> `this´.

Why just that one, though?  There are several more quotes in the
relevant Unicode block, why not add all of them?  They definitely fit
better for this task than ´, which is not even a quote character, and
isn't classified as punctuation.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 705 bytes --]


>> What we have no is ‘this’, 'this' and `this'.  What we could add is 
>> `this´.
>
> Why just that one, though?  There are several more quotes in the 
> relevant Unicode block, why not add all of them?
>

Indeed, all of them could be added.  With two options, then: 
text-left-quote-style and text-right-quote-style (and two similar options 
for buffers).

>
> They definitely fit better for this task than ´, which is not even a 
> quote character, and isn't classified as punctuation.
>

That depends on each one's preferences.  And ` is also not a quote 
character, it's a grave accent; to some at least it might seem natural to 
use an acute accent to match a grave accent.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Date: Fri, 17 Sep 2021 11:30:27 +0000
> From: Gregory Heytings <gregory@heytings.org>
> cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at, 
>     monnier@iro.umontreal.ca, dgutov@yandex.ru, larsi@gnus.org, 
>     juri@linkov.net
> 
> > They definitely fit better for this task than ´, which is not even a 
> > quote character, and isn't classified as punctuation.
> 
> That depends on each one's preferences.  And ` is also not a quote 
> character, it's a grave accent

Which is why we are trying to stop using it as a quote character.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 668 bytes --]


>>> They definitely fit better for this task than ´, which is not even a 
>>> quote character, and isn't classified as punctuation.
>>
>> That depends on each one's preferences.  And ` is also not a quote 
>> character, it's a grave accent
>
> Which is why we are trying to stop using it as a quote character.
>

Except that doing this throws the baby out with the bathwater.  If some 
think it looks ugly, it is IMO better to make it look better, instead of 
stopping to use it.  Because it's much easier to parse (as a pair) than 
double apostrophes.  It's a historical accident that ASCII included paired 
{} () [] <>, but no paired quotation marks.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Yuri Khan]

On Fri, 17 Sept 2021 at 19:32, Gregory Heytings <gregory@heytings.org> wrote:

> Except that doing this throws the baby out with the bathwater.  If some
> think it looks ugly, it is IMO better to make it look better, instead of
> stopping to use it.  Because it's much easier to parse (as a pair) than
> double apostrophes.  It's a historical accident that ASCII included paired
> {} () [] <>, but no paired quotation marks.

It’s a historical accident that the typewriter character set conflated
the opening single quote, closing single quote, apostrophe, acute
accent, and single prime. It is also a historical accident that the
grave accent ` was born as a separate spacing character at all — and
that is why I, among others, consider it ugly. Both these accidents
are canonically fixed by Unicode.

In my opinion, it is a mistake to avoid Unicode characters for the
reason that some terminals cannot display them. Teach those terminals
to map them to dumb quotes instead. Also, if Unicode quotation marks
are hard to enter, push for better keyboard layouts. (Personally, I
have all of “”‘’…–—«»°®©™≤≥§−±×÷≈ in my AltGr layer.)

Re: master ff4de1b: Fix quoting style in Lisp comments

[João Távora]

On Fri, Sep 17, 2021 at 5:23 PM Yuri Khan <yuri.v.khan@gmail.com> wrote:
>
> On Fri, 17 Sept 2021 at 19:32, Gregory Heytings <gregory@heytings.org> wrote:
>
> In my opinion, it is a mistake to avoid Unicode characters for the
> reason that some terminals cannot display them. Teach those terminals
> to map them to dumb quotes instead. Also, if Unicode quotation marks
> are hard to enter, push for better keyboard layouts. (Personally, I
> have all of “”‘’…–—«»°®©™≤≥§−±×÷≈ in my AltGr layer.)

No, don't push my programmer's muscle memory which wants
to work with the set of characters that 99% if not all the programming
languages use for identifiers and syntactic constructs.  At least
don't do it to me while I'm programming (elisp code, comments,
Git commit messages) where my brain needs as little Unicode
noise as possible.  Switching to the US layout where most language
designers designed popular programming languages in the second
half of the 20th century was one of the most productive decisions
of my career.

If I want to type type fancy smileys (or my own name João) in
non-programming Unicode capable places, that's another story.

João

Re: master ff4de1b: Fix quoting style in Lisp comments

[Alan Mackenzie]

On Fri, Sep 17, 2021 at 23:21:41 +0700, Yuri Khan wrote:
> On Fri, 17 Sept 2021 at 19:32, Gregory Heytings <gregory@heytings.org> wrote:

> > Except that doing this throws the baby out with the bathwater.  If
> > some think it looks ugly, it is IMO better to make it look better,
> > instead of stopping to use it.  Because it's much easier to parse
> > (as a pair) than double apostrophes.  It's a historical accident
> > that ASCII included paired {} () [] <>, but no paired quotation
> > marks.

> It’s a historical accident that the typewriter character set conflated
> the opening single quote, closing single quote, apostrophe, acute
> accent, and single prime. It is also a historical accident that the
> grave accent ` was born as a separate spacing character at all — and
> that is why I, among others, consider it ugly. Both these accidents
> are canonically fixed by Unicode.

Unicode is just a code.  To say it's a fix for the current problems is
to make a category error.  What you're really saying is that some
Unicode decorative (as opposed to functional) characters solve this
problem.  They don't, precisely because they're only decorative
characters.

> In my opinion, it is a mistake to avoid Unicode characters for the
> reason that some terminals cannot display them.

In my opinion, it's a good reason to avoid these impractical characters,
but not the most important one.

> Teach those terminals to map them to dumb quotes instead. Also, if
> Unicode quotation marks are hard to enter, push for better keyboard
> layouts.

This is the killer.  These decorative quote marks aren't merely hard to
enter, they're near to impossible on a qwerty (or similar) keyboard
layout.  This is a marked inconvenience to me when reading Info manuals.
I used to be able to search for the `marks' around symbols.  Since the
TexInfo people started using decorative characters, I can't do this any
more.

I can't type these characters.  More to the point, I can't refer to them
after other people have typed them.  Their introduction would signal a
loss in productivity and and increase in irritation for Emacs
developers.

The thing to alter, as Gregory said, is the appearance of the ASCII
characters on the terminals of those who find them annoying.

> (Personally, I have all of “”‘’…–—«»°®©™≤≥§−±×÷≈ in my AltGr layer.)

I only have characters there I need only rarely.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Yuri Khan <yuri.v.khan@gmail.com>
> Date: Fri, 17 Sep 2021 23:21:41 +0700
> Cc: Eli Zaretskii <eliz@gnu.org>, Stefan Kangas <stefan@marxist.se>, 
> 	Emacs developers <emacs-devel@gnu.org>, martin rudalics <rudalics@gmx.at>, 
> 	Stefan Monnier <monnier@iro.umontreal.ca>, Dmitry Gutov <dgutov@yandex.ru>, 
> 	Lars Magne Ingebrigtsen <larsi@gnus.org>, Juri Linkov <juri@linkov.net>
> 
> On Fri, 17 Sept 2021 at 19:32, Gregory Heytings <gregory@heytings.org> wrote:
> 
> In my opinion, it is a mistake to avoid Unicode characters for the
> reason that some terminals cannot display them. Teach those terminals
> to map them to dumb quotes instead.

That's only possible in Emacs, not in other programs, which we don't
control.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Yuri Khan]

On Sat, 18 Sept 2021 at 00:48, Eli Zaretskii <eliz@gnu.org> wrote:

> > In my opinion, it is a mistake to avoid Unicode characters for the
> > reason that some terminals cannot display them. Teach those terminals
> > to map them to dumb quotes instead.
>
> That's only possible in Emacs, not in other programs, which we don't
> control.

I suggested that as a feature of the affected terminals, not Emacs or programs.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Yuri Khan <yuri.v.khan@gmail.com>
> Date: Sat, 18 Sep 2021 01:42:51 +0700
> Cc: Stefan Kangas <stefan@marxist.se>, Emacs developers <emacs-devel@gnu.org>,
>  martin rudalics <rudalics@gmx.at>, Gregory Heytings <gregory@heytings.org>,
>  Stefan Monnier <monnier@iro.umontreal.ca>, Dmitry Gutov <dgutov@yandex.ru>,
>  Lars Magne Ingebrigtsen <larsi@gnus.org>, Juri Linkov <juri@linkov.net>
> 
> On Sat, 18 Sept 2021 at 00:48, Eli Zaretskii <eliz@gnu.org> wrote:
> 
> > > In my opinion, it is a mistake to avoid Unicode characters for the
> > > reason that some terminals cannot display them. Teach those terminals
> > > to map them to dumb quotes instead.
> >
> > That's only possible in Emacs, not in other programs, which we don't
> > control.
> 
> I suggested that as a feature of the affected terminals, not Emacs or programs.

Not possible with every terminal.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 1575 bytes --]


>> Except that doing this throws the baby out with the bathwater.  If some 
>> think it looks ugly, it is IMO better to make it look better, instead 
>> of stopping to use it.  Because it's much easier to parse (as a pair) 
>> than double apostrophes.  It's a historical accident that ASCII 
>> included paired {} () [] <>, but no paired quotation marks.
>
> It’s a historical accident that the typewriter character set conflated 
> the opening single quote, closing single quote, apostrophe, acute 
> accent, and single prime. It is also a historical accident that the 
> grave accent ` was born as a separate spacing character at all — and 
> that is why I, among others, consider it ugly. Both these accidents are 
> canonically fixed by Unicode.
>

The point is not to discuss whether Unicode makes a better distinction 
between different kinds of quotation marks and apostrophe.  It does, 
obviously.  But (1) these characters are difficult to enter on most 
keyboards, (2) it is not realistic to expect that keyboards will change, 
as 99.9% users do not need or even understand those subtleties, and word 
processors already do TRT when they press the " or ' keys, and (3) more 
importantly, using `foobar' is better than using any of these Unicode 
characters, because ` and ' are logical quotation marks or markups that 
can be displayed as any of the actual quotation marks depending on the 
user preferences.  Some will prefer ‛foobar’, others ‘foobar’, others 
‟foobar”, others “foobar”, others 'foobar', and so forth.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

> The point is not to discuss whether Unicode makes a better distinction
> between different kinds of quotation marks and apostrophe.  It does,
> obviously.  But (1) these characters are difficult to enter on most
> keyboards, (2) it is not realistic to expect that keyboards will change, as
> 99.9% users do not need or even understand those subtleties, and word

100% agreement.  My preference for using a well-defined convention based
on some existing markup language is that the text is then plain ASCII
(and hence easy to type and to display) but we can additionally prettify
it in a reliable way if the user cares about it.


        Stefan

Re: master ff4de1b: Fix quoting style in Lisp comments

[Manuel Giraud]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> The point is not to discuss whether Unicode makes a better distinction
>> between different kinds of quotation marks and apostrophe.  It does,
>> obviously.  But (1) these characters are difficult to enter on most
>> keyboards, (2) it is not realistic to expect that keyboards will change, as
>> 99.9% users do not need or even understand those subtleties, and word
>
> 100% agreement.  My preference for using a well-defined convention based
> on some existing markup language is that the text is then plain ASCII
> (and hence easy to type and to display) but we can additionally prettify
> it in a reliable way if the user cares about it.

[Hi! old time newbie here trying to follow this long discussion.]

I think that the point (3) of Gregory is important too.

It seems to me that you are refering to the `` Markdown convention. I
think that what Gregory was saying (correct me otherwise) is that it is
useful to have distinct characters to represent opening and closing
quote.
-- 
Manuel Giraud

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> I think that what Gregory was saying (correct me
> otherwise) is that it is useful to have distinct
> characters to represent opening and closing quote.

Several of us mentioned that advantage.  I know
I called it out - both in this thread and in the
infamous previous discussion(s).

Easier to recognize, fontify, search for.

(And easier than curly quotes to type, whether
for insertion or in a search pattern.)

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

>> 100% agreement.  My preference for using a well-defined convention based
>> on some existing markup language is that the text is then plain ASCII
>> (and hence easy to type and to display) but we can additionally prettify
>> it in a reliable way if the user cares about it.
>
> [Hi! old time newbie here trying to follow this long discussion.]
> I think that the point (3) of Gregory is important too.
> It seems to me that you are refering to the `` Markdown convention.

Not specifically, no.

> I think that what Gregory was saying (correct me otherwise) is that it
> is useful to have distinct characters to represent opening and
> closing quote.

The usefulness is that it resolves some ambiguities.  But those can
resolved in other ways (e.g. by disallowing newlines between the opening
and closing quote).  In any case, resolving ambiguity is why I'm
referring to a "well-defined convention".


        Stefan

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > it is useful to have distinct characters to
> > represent opening and closing quote.
> 
> The usefulness is that it resolves some ambiguities.
> But those can resolved in other ways (e.g. by
> disallowing newlines between the opening and closing
> quote).

Why would we want to impose such a restriction?

Why wouldn't we want the simplicity and ease of
different chars for opening & closing delimiters?

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

> Why would we want to impose such a restriction?

E.g. because multiline constructs are fundamentally harder to handle in
many cases, such as in font-lock?

Or because of things like

    ;; foo `bar
    ;; baz' toto

where the opener is not in the same comment as the closer.
In `typer-mode` I do allow multiline quoted constructs, but require
a different syntax for them (based on Markdown's four-space-indents).


        Stefan

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> In `typer-mode` I do allow multiline quoted constructs, but require
> a different syntax for them (based on Markdown's four-space-indents).

What's `typer-mode'?  I can't find it on GNU ELPA.

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

Stefan Kangas [2021-09-18 10:46:32] wrote:
> Stefan Monnier <monnier@iro.umontreal.ca> writes:
>> In `typer-mode` I do allow multiline quoted constructs, but require
>> a different syntax for them (based on Markdown's four-space-indents).
> What's `typer-mode'?  I can't find it on GNU ELPA.

Sorry, I ended up erasing the explanation while keeping the name  It's
just a mode for a local experimental language.


        Stefan

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > > > it is useful to have distinct characters to
> > > > represent opening and closing quote.
> > > 
> > > The usefulness is that it resolves some ambiguities.
> > > But those can resolved in other ways (e.g. by
> > > disallowing newlines between the opening and closing
> > > quote).
> >
> > Why would we want to impose such a restriction?
> 
> E.g. because multiline constructs are fundamentally
> harder to handle in many cases, such as in font-lock?

Sure, font-locking such things (but not only such
things) isn't 100% perfect.

Having distinct open and close delimiters makes
things easier not only for fontifying but also
for users.

> Or because of things like
>     ;; foo `bar
>     ;; baz' toto

Looks like a pretty concocted example.  Ever seen
that in the wild?

We don't currently font-lock even `...' with space
chars in the ..., in Lisp comments.  (I mentioned
earlier that we could.)

I do that in Info, for example (but emacs -Q
doesn't).

In the Emacs and Elisp manuals there are plenty of
occurrences of '...' where the ... contains space
chars - instances ranging from shell commands, such
as 'cp -r', to key bindings, such as 'a +', to font
names, such as 'song ti', to programming constructs,
such as 'else if', to VC commands, such as
'svn switch', to key sequences, such as 'M-x diary'
and '100 C-f', to key bindings, such as 'C-c C-c',
to patterns, such as 'C-x v r BRANCH-NAME <RET>'
and 'MONTHNAME DAY', to file extensions, such as
'.elc', to menu items, such as 'Other calendars'.

(Those are all from the Emacs manual, and there
are zillions more.)

But sure, it's more problematic to allow for
newline chars.  In my code (`info+.el') that's
optional, and not on by default (except for "...").
But it can by useful, to cover things like these:

'C-x
<RET>' 

'MONTHNAME DAY,
YEAR'

'ls
-l'

'xdvi
OUTPUT-FILE-NAME'

"newline
characters"

All of those occurrences, and many such more,
are in the Emacs manual, and they all fontify
nicely.  (And of course the same was true with
`...', before the change to curly quotes.)

I haven't done the same for `...' in Elisp
comments - only symbol constituents are
fontified there still.  But more could be done.

> where the opener is not in the same comment
> as the closer.

Yes, that would be a minor hiccup.  Fontifying
would need to disregard the comment-starts.
But as I say, that's a pretty uncommon case.

Font-locking need not aim to be perfect, even
in the default case.  But best is to make it
easy for users to switch among different
degrees of font-locking such things.

By default, `info+.el' fontifies all such
quotations.  For all except "..." (strings),
it does so only if the pattern is all on the
same line.  (By default, <...> text is also
fontified.)  Change the value of an option to
get multiline fontifying of all such.

You can also fontify isolated backquote and
quote chars, which is helpful for Lisp code
snippets (including in strings: "the `foo'").

It's easy to change such behavior on the fly.

Re: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Monnier]

> Having distinct open and close delimiters makes
> things easier not only for fontifying but also
> for users.

If you reread what I wrote, I never said I'm opposed to having different
delimiters for open and close.  Just that this part of the choice is not
terribly important IMO.

>> Or because of things like
>>     ;; foo `bar
>>     ;; baz' toto
>
> Looks like a pretty concocted example.  Ever seen
> that in the wild?

Huh?  Can you think of another example that spans multiple lines for all
those languages that use newline-terminated comments?


        Stefan

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: Yuri Khan <yuri.v.khan@gmail.com>,  Eli Zaretskii <eliz@gnu.org>,
>   Stefan Kangas <stefan@marxist.se>,  Emacs developers
>  <emacs-devel@gnu.org>,  martin rudalics <rudalics@gmx.at>,  Dmitry Gutov
>  <dgutov@yandex.ru>,  Lars Magne Ingebrigtsen <larsi@gnus.org>,  Juri
>  Linkov <juri@linkov.net>
> Date: Fri, 17 Sep 2021 15:22:12 -0400
> 
> > The point is not to discuss whether Unicode makes a better distinction
> > between different kinds of quotation marks and apostrophe.  It does,
> > obviously.  But (1) these characters are difficult to enter on most
> > keyboards, (2) it is not realistic to expect that keyboards will change, as
> > 99.9% users do not need or even understand those subtleties, and word
> 
> 100% agreement.  My preference for using a well-defined convention based
> on some existing markup language is that the text is then plain ASCII
> (and hence easy to type and to display) but we can additionally prettify
> it in a reliable way if the user cares about it.

The GNU preferences are described in standards.texi, and we should
follow them, not our personal preferences.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

>> 100% agreement.  My preference for using a well-defined convention based
>> on some existing markup language is that the text is then plain ASCII
>> (and hence easy to type and to display) but we can additionally prettify
>> it in a reliable way if the user cares about it.
>
> The GNU preferences are described in standards.texi, and we should
> follow them, not our personal preferences.

Does standards.texi offer advice on how to format Emacs Lisp comments,
or does it otherwise contradict anything that Stefan M wrote?

I skimmed `(standards) Documentation' but couldn't find anything.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Gregory Heytings]

[-- Attachment #1: Type: text/plain, Size: 1771 bytes --]


>>> My preference for using a well-defined convention based on some 
>>> existing markup language is that the text is then plain ASCII (and 
>>> hence easy to type and to display) but we can additionally prettify it 
>>> in a reliable way if the user cares about it.
>>
>> The GNU preferences are described in standards.texi, and we should 
>> follow them, not our personal preferences.
>
> Does standards.texi offer advice on how to format Emacs Lisp comments, 
> or does it otherwise contradict anything that Stefan M wrote?
>

AFAICS, it doesn't.  Here is the 5.10 Quote Characters section of the GCS:

In the C locale, the output of GNU programs should stick to plain ASCII 
for quotation characters in messages to users: preferably 0x22 (‘"’) or 
0x27 (‘'’) for both opening and closing quotes. Although GNU programs 
traditionally used 0x60 (‘`’) for opening and 0x27 (‘'’) for closing 
quotes, nowadays quotes ‘`like this'’ are typically rendered 
asymmetrically, so quoting ‘"like this"’ or ‘'like this'’ typically looks 
better.

It is ok, but not required, for GNU programs to generate locale-specific 
quotes in non-C locales. For example:

printf (gettext ("Processing file '%s'..."), file);

Here, a French translation might cause gettext to return the string 
"Traitement de fichier ‹ %s ›...", yielding quotes more appropriate for a 
French locale.

Sometimes a program may need to use opening and closing quotes directly. 
By convention, gettext translates the string ‘"`"’ to the opening quote 
and the string ‘"'"’ to the closing quote, and a program can use these 
translations. Generally, though, it is better to translate quote 
characters in the context of longer strings.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> > That depends on each one's preferences.  And ` is also not a quote
> > character, it's a grave accent
> 
> Which is why we are trying to stop using it as a quote character.

Really?  Earlier I think you said it was only
because it was considered by some (e.g. Paul)
to be ugly.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> This shows that Unicode quotes consistently look
> good, while the others look antiquated at best.  

Again with the newer-is-better refrain.
Antiquated, really?

And that means ugly?  Ugliness is in the
eye of the beholder.

> IMO, we should automatically display in comments and docstrings
> whatever quote character the user prefers according to
> `text-quoting-style', and automatically insert the correct
> quote character when typing "'".

So if someone prefers ` and types ' she gets
`?  So each time she types (setq foo 'toto)
she gets (setq foo `toto)?  Whether in a doc
string, a comment, or elsewhere?  Why would
that make sense?

Doing that anywhere would be awful, IMO.  But
the main (only?) case we're talking about is,
I think, Lisp comments.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> The main arguments were:
> 
>   . the ` character looks ugly on many modern displays
>   . the Unicode quotes are not universally supported
>     on text-mode terminals

As for the first argument: Ugliness is in the eye
of the beholder.  And that can depend on what one's
used to.

Yes, many people might not be used to `...'.
That's perhaps always been true, but even if not,
it's no doubt true now.

Many people new to Emacs are not used to lots of
things Emacs does.  Emacs is an eye-opener to many.
Likewise, Lisp.

It's not hard to get used to `...', IMO, and when
you do, you can see its simple beauty and utility,
including ease of recognition, search, fontifying,
and typing (versus typing curly quotes).

Not to mention identical appearance and behavior in
text-mode terminals and elsewhere.  No need for the
former to be an exception.  (There was no need to
switch to curly quotes for the latter, non-text-mode
case, just because some thought curly is less ugly.)

You'll no doubt say that that ship has sailed.  It
sailed with a not insignificant degree of opposition,
but it sailed.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 15.09.2021 23:15, Stefan Monnier wrote:
>> Then, if you don't mind, I will revert your commit.
> 
> I think we should first try to agree on a convention (specifically for
> quoting symbols in ELisp comments).  Either:
> 
> - We have no official preference (in which case the maintainer of
>    a package/file can try and impose his preference for that file, but
>    beyond this we should just refrain from changing other people's
>    style).  I guess that would mean you shouldn't revert his commit.

No, that would mean I should because he went and changed code I am 
responsible for from the style I prefer into a broken one.

> - We declare '...' as the official style and decide on how we deal with
>    the many cases of other style used throughout the code, and someone™
>    will have to write the new code for font-lock and completion.

That's my main point: doing this change now all of a sudden without the 
associated changes in infrastructure is bad form.

> - We declare `...' as the official style.  This matches the historical
>    use, so it requires no changes, AFAIK (other than reverting Eli's
>    commit, obviously).
> 
> - We declare something else as the official style.  I personally have
>    grown fond of (a subset of) Markdown, so I'd vote for `...` (and
>    encourage extending this to a few more conventions, e.g. so we can
>    distinguish code from prose in comments), but (a subset of) Org style
>    would also make a lot of sense.

`...` is fine with me, or anything else that looks like markup ('...' 
will result in more annoying false positives). But it will result in a 
lot more work across the board.

> This is a typical bikeshed kind of question, so there's not much point
> getting people to argue over it.  The maintainers should make an
> executive decision on it (but they do need to come to an agreement
> among themselves ;-).

And they need to consider the technical tradeoffs of every choice, and 
take responsibility for the necessary changes in supporting code.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: Eli Zaretskii <eliz@gnu.org>, Stefan Kangas <stefan@marxist.se>,
>  juri@linkov.net, rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 16 Sep 2021 13:50:32 +0300
> 
> > - We declare something else as the official style.  I personally have
> >    grown fond of (a subset of) Markdown, so I'd vote for `...` (and
> >    encourage extending this to a few more conventions, e.g. so we can
> >    distinguish code from prose in comments), but (a subset of) Org style
> >    would also make a lot of sense.
> 
> `...` is fine with me, or anything else that looks like markup ('...' 
> will result in more annoying false positives). But it will result in a 
> lot more work across the board.

Quoting `like this` flies in the face of the main reason why we moved
from `..'.  If `..` is okay, we might as well go back to the original
`..'.  I do wonder, though, where were these arguments back when
someone, I think Paul Eggert, said that ` looks ugly on modern
terminals, which triggered a lot of changes, in addition to this one.
Was all that in vain?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 14:39, Eli Zaretskii wrote:
>> Cc: Eli Zaretskii<eliz@gnu.org>, Stefan Kangas<stefan@marxist.se>,
>>   juri@linkov.net,rudalics@gmx.at,larsi@gnus.org,emacs-devel@gnu.org
>> From: Dmitry Gutov<dgutov@yandex.ru>
>> Date: Thu, 16 Sep 2021 13:50:32 +0300
>>
>>> - We declare something else as the official style.  I personally have
>>>     grown fond of (a subset of) Markdown, so I'd vote for `...` (and
>>>     encourage extending this to a few more conventions, e.g. so we can
>>>     distinguish code from prose in comments), but (a subset of) Org style
>>>     would also make a lot of sense.
>> `...` is fine with me, or anything else that looks like markup ('...'
>> will result in more annoying false positives). But it will result in a
>> lot more work across the board.
> Quoting `like this` flies in the face of the main reason why we moved
> from `..'.  If `..` is okay, we might as well go back to the original
> `..'.  I do wonder, though, where were these arguments back when
> someone, I think Paul Eggert, said that ` looks ugly on modern
> terminals, which triggered a lot of changes, in addition to this one.
> Was all that in vain?

We did enact some changes which weren't disputed by anybody. For 
example, the display of quotes in messages, errors and Help buffers.

Even if we might perhaps switch to a different way of showing those 
quotes in Help buffers later (e.g. hide the quote chars and rely on 
different face).

Regarding the Markdown style, it might be a matter of interpreting 
feedback. IIRC the core of Paul's argument was that his students said 
`..' quoting style appeared old/outdated (basically it looked alien).

But Markdown for source files is quite popular, so it stands to reason 
that PL students might look at it differently. Or that perhaps the 
backquote itself was not the problem, at least not in source files.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> We did enact some changes which weren't disputed by anybody. For
> example, the display of quotes in messages, errors and Help buffers.

Those changes (to curly quotes) were disputed by me.
I'm somebody, even if only one body.

And I wasn't the only one arguing against changing
to curly quotes in messages and Help buffers.

> Even if we might perhaps switch to a different
> way of showing those quotes in Help buffers later
> (e.g. hide the quote chars and rely on different face).

Please don't do that.

> IIRC the core of Paul's argument was that his
> students said `..' quoting style appeared
> old/outdated (basically it looked alien).

(Did anyone actually hear from the students?)
But sure.  Lots of stuff in Emacs looks alien
at first.  A tractor looks alien to someone
who's only used a horse to plough.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> Quoting `like this` flies in the face of the main reason why we moved
> from `..'.  If `..` is okay, we might as well go back to the original
> `..'.

Yes, please do.  The sooner the better.

> I do wonder, though, where were these arguments back when
> someone, I think Paul Eggert, said that ` looks ugly on modern
> terminals, which triggered a lot of changes, in addition to this one.
> Was all that in vain?

"These arguments" _were_ expressed at the time,
clearly and with vigor.

The "ugly" cosmetics view `...' by Paul was
opposed by more than one (I'm only one, of
course).  Paul won the day with your decision.

And yes, "all that" was "in vain".  Vanity
misguided.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Stefan Kangas]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> - We declare something else as the official style.  I personally have
>   grown fond of (a subset of) Markdown, so I'd vote for `...` (and
>   encourage extending this to a few more conventions, e.g. so we can
>   distinguish code from prose in comments), but (a subset of) Org style
>   would also make a lot of sense.

BTW, Org-mode has a very nice optional feature where you don't even
see the ugly bits: when you type "**foo**", the stars disappear and
"foo" is displayed in italics.  If the problem is that some people
find the ` character ugly, perhaps we could have something similar for
our comments and/or docstrings to have ` and ' displayed as the proper
unicode quote marks.

It would be even better if we could have the ' key automatically
balance quotes, much like what happens in modern word processors.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > - We declare `...' as the official style.  This matches the historical
  >   use, so it requires no changes, AFAIK (other than reverting Eli's
  >   commit, obviously).

That IS the official style, in Emacs Lisp files, and we should not
change that.  We should use `...' in comments.

If the change make Font-Lock mode highlight '...' like `...' in Emacs
Lisp mode, that isn't necessarily a bad change.  It would highlight in
a useful way some code that doesn't follow our official style.  That
in itself is an improvement, all else being equal.

Would the change give wrong results in some cases?  Maybe.  I think
that depends on the details of in which cases it recognizes '...'.
If the cases are narrowly limited, maybe it won't give wrong results
very much.

Text files (whether Text mode or some other mode) are a different
issue.  Maybe it is ok to change to quote symbols in them with '..'.
I won't argue against that.

Still, I think we should make highlighting give correct results for
files that use `...' as well as for files that use '...'.  That should
be possible if the ... stands for a limited set of possible contents.

-- 
Dr Richard Stallman (https://stallman.org)
Chief GNUisance of the GNU Project (https://gnu.org)
Founder, Free Software Foundation (https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)

Re: master ff4de1b: Fix quoting style in Lisp comments

[Alexandre Garreau]

Le vendredi 17 septembre 2021, 02:16:53 CEST Richard Stallman a écrit :
> Still, I think we should make highlighting give correct results for
> files that use `...' as well as for files that use '...'.  That should
> be possible if the ... stands for a limited set of possible contents.

That’s trivial given that a well-formed quotation doesn’t begin or end 
with a space or quotation symbol (in proper traditional english (but if we 
do it “the logical way” it *may* *end* with a punctuation symbol)), given 
that it will also normally follow a blank and be followed by a blank or a 
quotation symbol, that alone is sufficient.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> Cc: dgutov@yandex.ru, eliz@gnu.org, stefan@marxist.se,
> 	juri@linkov.net, rudalics@gmx.at, larsi@gnus.org,
> 	emacs-devel@gnu.org
> Date: Thu, 16 Sep 2021 20:16:53 -0400
> 
>   > - We declare `...' as the official style.  This matches the historical
>   >   use, so it requires no changes, AFAIK (other than reverting Eli's
>   >   commit, obviously).
> 
> That IS the official style, in Emacs Lisp files, and we should not
> change that.  We should use `...' in comments.

We've decided to change that style.  It makes no sense for a project
to back out its decisions so quickly, when the only reason is that
some people don't like the decision.  They didn't like it when the
decision was made, either (heck, I didn't like it too much as well),
so there's nothing new here that requires to go back.  We are past
that decision point, and many texts were meanwhile written using this
style.

> Text files (whether Text mode or some other mode) are a different
> issue.

No, they aren't.  When I write plain text, I don't want to remember
what file I'm writing it in, and neither should anyone else.

Again, we've had this discussion.  It was a long and painful one, and
we emerged from it with a decision.  If we want to be a functional
project, we should follow our own decisions, even if they are
sometimes unusual or require re-learning some simple stuff.  We have
many examples of such decisions, this is just one of them.  If people
are going to revolt because they disagree with decisions like that,
where will that take us?  E.g., I dislike the decision to remove
facemenu and its useful (for me) key bindings: should I now start a
lobbying thread explaining time and again why that should be reverted?
There's no end to this, and it is not a reasonable way to manage the
project development.

> Still, I think we should make highlighting give correct results for
> files that use `...' as well as for files that use '...'.

That is already so, primarily because we are not going to convert the
quoting style en masse any time soon.  The issue at hand is what to
use for NEW comments and comments that we modify anyway.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 17.09.2021 09:34, Eli Zaretskii wrote:
> E.g., I dislike the decision to remove
> facemenu and its useful (for me) key bindings: should I now start a
> lobbying thread explaining time and again why that should be reverted?
> There's no end to this, and it is not a reasonable way to manage the
> project development.

If people come out en masse and say they want the face menu back, we 
should seriously consider it.

Just like with any other change, especially those that are relatively 
easy to undo.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: monnier@iro.umontreal.ca, stefan@marxist.se, juri@linkov.net,
>  rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Fri, 17 Sep 2021 13:44:34 +0300
> 
> On 17.09.2021 09:34, Eli Zaretskii wrote:
> > E.g., I dislike the decision to remove
> > facemenu and its useful (for me) key bindings: should I now start a
> > lobbying thread explaining time and again why that should be reverted?
> > There's no end to this, and it is not a reasonable way to manage the
> > project development.
> 
> If people come out en masse and say they want the face menu back, we 
> should seriously consider it.

How many are "en masse"?  We have less than a dozen people claiming
they want `..' back, that's hardly a "masse" in my book.

> Just like with any other change, especially those that are relatively 
> easy to undo.

This one isn't, because we have been using it for many moons now.
Some of it is practically impossible to undo, because it's in Git
commit log messages.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 17.09.2021 14:11, Eli Zaretskii wrote:
>> Cc: monnier@iro.umontreal.ca, stefan@marxist.se, juri@linkov.net,
>>   rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
>> From: Dmitry Gutov <dgutov@yandex.ru>
>> Date: Fri, 17 Sep 2021 13:44:34 +0300
>>
>> On 17.09.2021 09:34, Eli Zaretskii wrote:
>>> E.g., I dislike the decision to remove
>>> facemenu and its useful (for me) key bindings: should I now start a
>>> lobbying thread explaining time and again why that should be reverted?
>>> There's no end to this, and it is not a reasonable way to manage the
>>> project development.
>>
>> If people come out en masse and say they want the face menu back, we
>> should seriously consider it.
> 
> How many are "en masse"?  We have less than a dozen people claiming
> they want `..' back, that's hardly a "masse" in my book.

I don't see anybody concur with your stance in this particular 
discussion, and if you search the old one, I think it was mostly Paul 
(?), and even he was really arguing toward something different, in a 
different context as well (plain text files).

Unlike some other questions, this is about the development process, and 
the team of regulars is not so big. A dozen people seems like a 
significant number for that purpose.

>> Just like with any other change, especially those that are relatively
>> easy to undo.
> 
> This one isn't, because we have been using it for many moons now.
> Some of it is practically impossible to undo, because it's in Git
> commit log messages.

Again, look at this discussion: nobody is disputing the format of log 
messages.

Lars said we shouldn't care about it too much, but that's neither here 
nor there.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: rms@gnu.org, monnier@iro.umontreal.ca, stefan@marxist.se,
>  juri@linkov.net, rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Fri, 17 Sep 2021 14:25:31 +0300
> 
> > How many are "en masse"?  We have less than a dozen people claiming
> > they want `..' back, that's hardly a "masse" in my book.
> 
> I don't see anybody concur with your stance in this particular 
> discussion, and if you search the old one, I think it was mostly Paul 
> (?), and even he was really arguing toward something different, in a 
> different context as well (plain text files).

It doesn't matter.  This is a decision we already made, so just saying
you don't like it is not good enough.  We need a much stronger reason
to reverse previous decisions.

> > This one isn't, because we have been using it for many moons now.
> > Some of it is practically impossible to undo, because it's in Git
> > commit log messages.
> 
> Again, look at this discussion: nobody is disputing the format of log 
> messages.

Not true.  If you look at the commit log messages, even the latest
ones done today, you will see that almost everyone disputes that part.

> Lars said we shouldn't care about it too much, but that's neither here 
> nor there.

Which is a clear disagreement with what I said, and contradicts what
you say above.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Phil Sainty]

On 2021-09-18 00:09, Eli Zaretskii wrote:
>> I don't see anybody concur with your stance in this particular
>> discussion, and if you search the old one, I think it was mostly Paul
>> (?), and even he was really arguing toward something different, in a
>> different context as well (plain text files).
> 
> It doesn't matter.  This is a decision we already made, so just saying
> you don't like it is not good enough.  We need a much stronger reason
> to reverse previous decisions.

I might have missed something, but I think almost everyone else in
this thread has expressed some degree of surprise/confusion that
elisp comments would be categorised as "plain text" and/or the notion
that the previous decision was ever intended to apply to those.

If the only strong reason for making a change here is to adhere to
a decision that nobody (or almost nobody) was *actually* trying to
make (where elisp comments are concerned), then that would seem like
a very strong reason to stop and reassess.

I don't know the answer, but I'm seeing a lot of confusion.

Can we at minimum get confirmation from the people who were arguing
for fancy quotes in those old discussions that they wanted *this*
as well?  Because if those people weren't trying to change elisp
comments, then I don't think the old decision is relevant here, in
which case nothing would be "reversed" by maintaining the traditional
quoting style.


-Phil

Re: master ff4de1b: Fix quoting style in Lisp comments

[Tassilo Horn]

Phil Sainty <psainty@orcon.net.nz> writes:

>> It doesn't matter.  This is a decision we already made, so just
>> saying you don't like it is not good enough.  We need a much stronger
>> reason to reverse previous decisions.
>
> I might have missed something, but I think almost everyone else in
> this thread has expressed some degree of surprise/confusion that elisp
> comments would be categorised as "plain text" and/or the notion that
> the previous decision was ever intended to apply to those.

Just to add another data point: I didn't know the new convention and
always used `...' in elisp code comments because it felt like the right
way simply because it got highlighted and I had symbol-completion
enabled after typing the ` in a comment.  And I wouldn't have considered
code comments in elisp files to be in the came category of plain-text
documents like NEWS or ChangeLog.

I don't consider `...' to be ugly but in the end I don't care as long as
the style going to be used for quoting symbols in code comments is
supported by font-lock and completion.  In his regard, I do consider
`...' or the markdown-style `...` to be quite good because they are
seldomly used in prose or for emphasis in contrast to '...' reducing the
likelyhood of false-positives.

Bye,
Tassilo

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Date: Sat, 18 Sep 2021 00:38:50 +1200
> From: Phil Sainty <psainty@orcon.net.nz>
> Cc: Dmitry Gutov <dgutov@yandex.ru>, rms@gnu.org, stefan@marxist.se,
>  juri@linkov.net, rudalics@gmx.at, monnier@iro.umontreal.ca, larsi@gnus.org,
>  emacs-devel@gnu.org
> 
> Can we at minimum get confirmation from the people who were arguing
> for fancy quotes in those old discussions that they wanted *this*
> as well?  Because if those people weren't trying to change elisp
> comments, then I don't think the old decision is relevant here, in
> which case nothing would be "reversed" by maintaining the traditional
> quoting style.

People are arguing not only about Lisp comments, they are also arguing
about Git commit log messages, using other quoting styles, and using
`..' everywhere.  So I no longer have a clear idea what this is about.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 17.09.2021 15:09, Eli Zaretskii wrote:
>> Cc: rms@gnu.org, monnier@iro.umontreal.ca, stefan@marxist.se,
>>   juri@linkov.net, rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
>> From: Dmitry Gutov <dgutov@yandex.ru>
>> Date: Fri, 17 Sep 2021 14:25:31 +0300
>>
>>> How many are "en masse"?  We have less than a dozen people claiming
>>> they want `..' back, that's hardly a "masse" in my book.
>>
>> I don't see anybody concur with your stance in this particular
>> discussion, and if you search the old one, I think it was mostly Paul
>> (?), and even he was really arguing toward something different, in a
>> different context as well (plain text files).
> 
> It doesn't matter.  This is a decision we already made, so just saying
> you don't like it is not good enough.  We need a much stronger reason
> to reverse previous decisions.

The cost of backtracking is low, so yes, if people even simply say "we 
don't like it", it should be changed.

I've listed particular reasons as well, but what does it matter: you 
ignore those parts of messages.

>>> This one isn't, because we have been using it for many moons now.
>>> Some of it is practically impossible to undo, because it's in Git
>>> commit log messages.
>>
>> Again, look at this discussion: nobody is disputing the format of log
>> messages.
> 
> Not true.

Nobody in *this discussion*. People being casually careless is nothing 
new, OTOH.

> If you look at the commit log messages, even the latest
> ones done today, you will see that almost everyone disputes that part.

This is orthogonal to this discussion. Please don't derail it.

Re: master ff4de1b: Fix quoting style in Lisp comments

[João Távora]

[-- Attachment #1: Type: text/plain, Size: 2088 bytes --]

Hopefully I'm not too late in adding my voice to those who want to keep the
traditional `style' in lisp comments, too.

I don't know if anyone mentioned it yet, but besides being font-locked,
it's autopaired by default in elisp-mode with electric-pair-mode, which is
nice, IMO.

João

On Fri, Sep 17, 2021, 14:08 Dmitry Gutov <dgutov@yandex.ru> wrote:

> On 17.09.2021 15:09, Eli Zaretskii wrote:
> >> Cc: rms@gnu.org, monnier@iro.umontreal.ca, stefan@marxist.se,
> >>   juri@linkov.net, rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
> >> From: Dmitry Gutov <dgutov@yandex.ru>
> >> Date: Fri, 17 Sep 2021 14:25:31 +0300
> >>
> >>> How many are "en masse"?  We have less than a dozen people claiming
> >>> they want `..' back, that's hardly a "masse" in my book.
> >>
> >> I don't see anybody concur with your stance in this particular
> >> discussion, and if you search the old one, I think it was mostly Paul
> >> (?), and even he was really arguing toward something different, in a
> >> different context as well (plain text files).
> >
> > It doesn't matter.  This is a decision we already made, so just saying
> > you don't like it is not good enough.  We need a much stronger reason
> > to reverse previous decisions.
>
> The cost of backtracking is low, so yes, if people even simply say "we
> don't like it", it should be changed.
>
> I've listed particular reasons as well, but what does it matter: you
> ignore those parts of messages.
>
> >>> This one isn't, because we have been using it for many moons now.
> >>> Some of it is practically impossible to undo, because it's in Git
> >>> commit log messages.
> >>
> >> Again, look at this discussion: nobody is disputing the format of log
> >> messages.
> >
> > Not true.
>
> Nobody in *this discussion*. People being casually careless is nothing
> new, OTOH.
>
> > If you look at the commit log messages, even the latest
> > ones done today, you will see that almost everyone disputes that part.
>
> This is orthogonal to this discussion. Please don't derail it.
>
>

[-- Attachment #2: Type: text/html, Size: 3333 bytes --]

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: rms@gnu.org, monnier@iro.umontreal.ca, stefan@marxist.se,
>  juri@linkov.net, rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Fri, 17 Sep 2021 16:07:17 +0300
> 
> > It doesn't matter.  This is a decision we already made, so just saying
> > you don't like it is not good enough.  We need a much stronger reason
> > to reverse previous decisions.
> 
> The cost of backtracking is low, so yes, if people even simply say "we 
> don't like it", it should be changed.

There are a lot of low-cost changes we could do, but I very much hope
we won't do them just because they are low-cost.

> I've listed particular reasons as well, but what does it matter: you 
> ignore those parts of messages.

Because you aren't saying anything new.

> > If you look at the commit log messages, even the latest
> > ones done today, you will see that almost everyone disputes that part.
> 
> This is orthogonal to this discussion.

It is?  Says who?  AFAICS, it is very much part of the discussion,
just read the messages.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 17.09.2021 18:49, Eli Zaretskii wrote:
>>> If you look at the commit log messages, even the latest
>>> ones done today, you will see that almost everyone disputes that part.
>> This is orthogonal to this discussion.
> It is?  Says who?  AFAICS, it is very much part of the discussion,
> just read the messages.

I started this thread, so it might be natural to ask my opinion.

Or you can just split this into several questions, to be considered 
separately:

1. Lisp comments; possibly also comments in .c files
2. Commit messages
3. Plain text files

FTR, I'm only talking about (1) here.

We don't have any special support for highlighting or otherwise working 
with symbol references in (2) and (3), so '...' syntax doesn't hurt.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> I've listed particular reasons as well, but what does it matter: you
> ignore those parts of messages.

I too listed several good reasons.

People have not just said, "I don't like it."

Re: master ff4de1b: Fix quoting style in Lisp comments

[Amin Bandali]

Dmitry Gutov writes:

> On 17.09.2021 14:11, Eli Zaretskii wrote:
>>> Cc: monnier@iro.umontreal.ca, stefan@marxist.se, juri@linkov.net,
>>>   rudalics@gmx.at, larsi@gnus.org, emacs-devel@gnu.org
>>> From: Dmitry Gutov <dgutov@yandex.ru>
>>> Date: Fri, 17 Sep 2021 13:44:34 +0300
>>>
>>> On 17.09.2021 09:34, Eli Zaretskii wrote:
>>>> E.g., I dislike the decision to remove
>>>> facemenu and its useful (for me) key bindings: should I now start a
>>>> lobbying thread explaining time and again why that should be reverted?
>>>> There's no end to this, and it is not a reasonable way to manage the
>>>> project development.
>>>
>>> If people come out en masse and say they want the face menu back, we
>>> should seriously consider it.
>> How many are "en masse"?  We have less than a dozen people claiming
>> they want `..' back, that's hardly a "masse" in my book.
>
> I don't see anybody concur with your stance in this particular
> discussion, and if you search the old one, I think it was mostly Paul
> (?), and even he was really arguing toward something different, in a
> different context as well (plain text files).
>
> Unlike some other questions, this is about the development process,
> and the team of regulars is not so big. A dozen people seems like a
> significant number for that purpose.

I acknowledge that this ship seems to have sailed (though my personal
preference would be a reconsideration of the choice, and it seems like
several other regulars here want that as well) and I may not have a
say in it now (the long discussion in 2015 predates my involvement
in/around Emacs and emacs-devel), but if I may voice my opinion now
that I /am/ around and somewhat active, I prefer `...' over '...'.
It's a quoting style familiar from several other GNU projects I have
been around and/or use -- e.g. M4, Guile, ... -- and to me it's always
been a welcome uniqueness (if one may call it that) compared to the
rest of the software world, and I enjoy seeing it in docstrings,
comments, changelogs, and so on.

I've used a large number of fonts over the years, and don't see what
the supposed 'ugliness' of the ` is (as others have pointed out, it's
already used by various markups like the popular Markdown, and doesn't
seem to be problematic there).  For pretty-printed documentation and
manuals it's nice that ‘...’ unicode quotes are used for display, but
as mentioned they are not supported everywhere and are not as easy to
type.  I believe `...' is a nice ASCII dual of that, one which we've
already been using in Emacs for years.  I find that the opening and
closing characters being different gives me a pleasant, visual cue
about the start and end point of a quote.  Sure, it may be different
to what programmers are typically used to, but I don't think that
makes it any less useful or would be justification enough to throw it
out the window.

In short, I really like this ASCII typographic `hack' and find it
useful; and I like several of the points other folks -- like Phil --
have made in favour of not abandoning it. :)

>>> Just like with any other change, especially those that are relatively
>>> easy to undo.
>> This one isn't, because we have been using it for many moons now.
>> Some of it is practically impossible to undo, because it's in Git
>> commit log messages.
>
> Again, look at this discussion: nobody is disputing the format of log
> messages.
>
> Lars said we shouldn't care about it too much, but that's neither here
> nor there.

Yeah I don't think we'd need to 'undo' anything in the commit logs.
Those are there to stay, but much could be done in etc/NEWS and
elsewhere, and in future commit messages.  Also, as folks mentioned,
it's not like people have been exclusively using '...' in commit
messages or plain text files, and `...' is still used quite frequently
(and has been, historically), and I foresee that it will continue to
be.  Considering the long history of development of Emacs, I don't
think it's at all too late to reverse this decision if we'd like to.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> If people come out en masse and say they want the face menu back, we
> should seriously consider it.
> 
> Just like with any other change, especially those that are relatively
> easy to undo.

Please add 1 to the pending en masse.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: juri@linkov.net, rudalics@gmx.at, monnier@iro.umontreal.ca,
>  larsi@gnus.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Wed, 15 Sep 2021 23:01:40 +0300
> 
> On 15.09.2021 21:52, Eli Zaretskii wrote:
> >> From: Stefan Kangas<stefan@marxist.se>
> >> Date: Wed, 15 Sep 2021 20:40:06 +0200
> >> Cc: Juri Linkov<juri@linkov.net>, martin rudalics<rudalics@gmx.at>,
> >>   Stefan Monnier<monnier@iro.umontreal.ca>, Dmitry Gutov<dgutov@yandex.ru>,
> >>   Lars Ingebrigtsen<larsi@gnus.org>, Emacs developers<emacs-devel@gnu.org>
> >>
> >> Eli Zaretskii<eliz@gnu.org>  writes:
> >>
> >>>> Your commit broke symbol highlighting in the comments.
> >>> And `this` doesn't?
> >> Yes, it does.  That's why I originally changed it to `this'.
> > yes, everyone seems to do what they like in this respect.
> 
> One is not like the other.

How's that?

> > So I did as
> > well (except that in my case I have a discussion with a conclusion to
> > back me up, not that it matters).
> 
> Then, if you don't mind, I will revert your commit.

I do mind, of course.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 07:53, Eli Zaretskii wrote:
>>>>>> Your commit broke symbol highlighting in the comments.
>>>>> And `this` doesn't?
>>>> Yes, it does.  That's why I originally changed it to `this'.
>>> yes, everyone seems to do what they like in this respect.
>> One is not like the other.
> How's that?

One is working WRT to coding features, the other is broken.

>>> So I did as
>>> well (except that in my case I have a discussion with a conclusion to
>>> back me up, not that it matters).
>> Then, if you don't mind, I will revert your commit.
> I do mind, of course.

Then your claim about everybody doing what they want is void.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 16 Sep 2021 11:51:55 +0300
> Cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at,
>  monnier@iro.umontreal.ca, larsi@gnus.org, juri@linkov.net
> 
> On 16.09.2021 07:53, Eli Zaretskii wrote:
> >>>>>> Your commit broke symbol highlighting in the comments.
> >>>>> And `this` doesn't?
> >>>> Yes, it does.  That's why I originally changed it to `this'.
> >>> yes, everyone seems to do what they like in this respect.
> >> One is not like the other.
> > How's that?
> 
> One is working WRT to coding features, the other is broken.

We aren't talking about code.

> >>> So I did as
> >>> well (except that in my case I have a discussion with a conclusion to
> >>> back me up, not that it matters).
> >> Then, if you don't mind, I will revert your commit.
> > I do mind, of course.
> 
> Then your claim about everybody doing what they want is void.

No, it isn't.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 12:11, Eli Zaretskii wrote:
>> From: Dmitry Gutov <dgutov@yandex.ru>
>> Date: Thu, 16 Sep 2021 11:51:55 +0300
>> Cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at,
>>   monnier@iro.umontreal.ca, larsi@gnus.org, juri@linkov.net
>>
>> On 16.09.2021 07:53, Eli Zaretskii wrote:
>>>>>>>> Your commit broke symbol highlighting in the comments.
>>>>>>> And `this` doesn't?
>>>>>> Yes, it does.  That's why I originally changed it to `this'.
>>>>> yes, everyone seems to do what they like in this respect.
>>>> One is not like the other.
>>> How's that?
>>
>> One is working WRT to coding features, the other is broken.
> 
> We aren't talking about code.

As others have attested in this thread, people do not think of code 
comments as simply 'plain text'. Unlike NEWS, ChangeLogs, etc, which 
*have* been the subject of some previous discussion. If people had 
actually brought up comments in Lisp code, it would have been a much 
difficult discussion.

We have used the same markup in Lisp comments as in Lisp docstrings for 
symbol references for ages. With associated benefits in highlighting and 
code completion. Why would we suddenly change that?

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: stefan@marxist.se, emacs-devel@gnu.org, rudalics@gmx.at,
>  monnier@iro.umontreal.ca, larsi@gnus.org, juri@linkov.net
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 16 Sep 2021 13:34:27 +0300
> 
> We have used the same markup in Lisp comments as in Lisp docstrings for 
> symbol references for ages. With associated benefits in highlighting and 
> code completion. Why would we suddenly change that?

That ship has sailed when we had this discussion back when we had it.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 16.09.2021 14:35, Eli Zaretskii wrote:
>> Cc:stefan@marxist.se,emacs-devel@gnu.org,rudalics@gmx.at,
>>   monnier@iro.umontreal.ca,larsi@gnus.org,juri@linkov.net
>> From: Dmitry Gutov<dgutov@yandex.ru>
>> Date: Thu, 16 Sep 2021 13:34:27 +0300
>>
>> We have used the same markup in Lisp comments as in Lisp docstrings for
>> symbol references for ages. With associated benefits in highlighting and
>> code completion. Why would we suddenly change that?
> That ship has sailed when we had this discussion back when we had it.

We didn't have this discussion, we had a different one.

Re: master ff4de1b: Fix quoting style in Lisp comments

[martin rudalics]

 > I don't really see why people keep arguing, since everyone around here
 > uses the quoting they like anyway, disregarding the guidelines.  I see
 > `..', '..', and `..` all the time.  It's anarchy anyway, so why people
 > insist on arguing about something they won't follow anyway?  Just let
 > it go.

I would like to follow the guidelines but I don't know where to find them.

martin

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: larsi@gnus.org, dgutov@yandex.ru, monnier@iro.umontreal.ca,
>  emacs-devel@gnu.org
> From: martin rudalics <rudalics@gmx.at>
> Date: Wed, 15 Sep 2021 20:48:45 +0200
> 
>  > I don't really see why people keep arguing, since everyone around here
>  > uses the quoting they like anyway, disregarding the guidelines.  I see
>  > `..', '..', and `..` all the time.  It's anarchy anyway, so why people
>  > insist on arguing about something they won't follow anyway?  Just let
>  > it go.
> 
> I would like to follow the guidelines but I don't know where to find them.

Why do you need to find them?  I just described them up-thread.  They
are very simple.

Re: master ff4de1b: Fix quoting style in Lisp comments

[martin rudalics]

 >>   > I don't really see why people keep arguing, since everyone around here
 >>   > uses the quoting they like anyway, disregarding the guidelines.  I see
 >>   > `..', '..', and `..` all the time.  It's anarchy anyway, so why people
 >>   > insist on arguing about something they won't follow anyway?  Just let
 >>   > it go.
 >>
 >> I would like to follow the guidelines but I don't know where to find them.
 >
 > Why do you need to find them?  I just described them up-thread.  They
 > are very simple.

I suppose you refer to your earlier

 >> I'm not sure we ever agreed to standardize on straight quotes in Lisp
 >> comments (as opposed to plain text files like NEWS).
 >
 > Comments, NEWS, PROBLEMS, DEBUG, commit log messages (thus the
 > generated ChangeLog files), and any other form of plain-text
 > documentation.

and while I do not have the slightest problem with most of these I do
not understand what should happen to code I comment out.  If code I
intend to comment out contains a `...' should I automatically convert it
to '...' or should I leave it alone?

I'd prefer a sentence like

   Comments (excluding doc-strings written as comments and code that has
   been "commented out"), NEWS, PROBLEMS, DEBUG, commit log messages
   (thus the generated ChangeLog files), and any other form of plain-text
   documentation.

In the past I've been always trying hard to avoid the "anarchy" you
mention above and I still would like to get it right in the things I
write.

martin

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: larsi@gnus.org, dgutov@yandex.ru, emacs-devel@gnu.org,
>  monnier@iro.umontreal.ca, juri@linkov.net
> From: martin rudalics <rudalics@gmx.at>
> Date: Thu, 16 Sep 2021 08:56:42 +0200
> 
>  > Comments, NEWS, PROBLEMS, DEBUG, commit log messages (thus the
>  > generated ChangeLog files), and any other form of plain-text
>  > documentation.
> 
> and while I do not have the slightest problem with most of these I do
> not understand what should happen to code I comment out.  If code I
> intend to comment out contains a `...' should I automatically convert it
> to '...' or should I leave it alone?

My opinion is that if it's code that one day will be uncommented, it's
okay to leave the `...' alone.  But if we are sure this will likely
always be a comment, then yes, a conversion would be best.

> In the past I've been always trying hard to avoid the "anarchy" you
> mention above and I still would like to get it right in the things I
> write.

If people will follow the rules, it is indeed worth our while to
discuss them and make them detailed enough.

RE: [External] : Re: master ff4de1b: Fix quoting style in Lisp comments

[Drew Adams]

> My opinion is that if it's code that one day will be uncommented, it's
> okay to leave the `...' alone.  But if we are sure this will likely
> always be a comment, then yes, a conversion would be best.

If you're including Lisp comments in that, then I,
for one, have a different opinion.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 13.09.2021 14:34, Eli Zaretskii wrote:
>> From: Dmitry Gutov <dgutov@yandex.ru>
>> Date: Mon, 13 Sep 2021 01:42:08 +0300
>>
>> Hi Eli,
>>
>> On 12.09.2021 19:51, Eli Zaretskii wrote:
>>> -;; `eval' is hence the core of the culprit.  It's used on:
>>> +;; 'eval' is hence the core of the culprit.  It's used on:
>>>
>>> -;; encouraged to use `lisp-data-mode' instead.
>>> +;; encouraged to use 'lisp-data-mode' instead.
>>>
>>> -;; The time- functions below translate nil to `current-time' and
>>> -;; accept an integer as of Emacs 25.  `decode-time' and
>>> -;; `format-time-string' accept nil on Emacs 24 but don't accept an
>>> +;; The time- functions below translate nil to 'current-time' and
>>> +;; accept an integer as of Emacs 25.  'decode-time' and
>>> +;; 'format-time-string' accept nil on Emacs 24 but don't accept an
>>
>> Is this a good idea?
> 
> We decided to use this style in plain-text documentation long ago.

IME code comments are not the same as plain-text documentation.

Aside from highlighting, we have other code that relies on unambiguous 
markup. Emacs Lisp completion code also enters "symbol completion" mode 
(so to speak) after you type '`', allowing one to more easily input an 
existing symbol.

The fact that both font-lock and completion in Elisp do not recognize 
'...' currently is I think a good indicator that we have not really made 
a choice regarding this syntax in comments yet, and can reconsider.

>> I'm not sure we ever agreed to standardize on straight quotes in Lisp
>> comments (as opposed to plain text files like NEWS).
> 
> Comments, NEWS, PROBLEMS, DEBUG, commit log messages (thus the
> generated ChangeLog files), and any other form of plain-text
> documentation.

Having docstring markup in plain comments is fairly common across 
languages, I think.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Dmitry Gutov <dgutov@yandex.ru> writes:

> The fact that both font-lock and completion in Elisp do not recognize
> '...' currently is I think a good indicator that we have not really
> made a choice regarding this syntax in comments yet, and can
> reconsider.

Yup.  And as a practical matter, a quick grep shows that we have more
than 10K lines that match ";;.*`", so the code churn would be huge.  So
changing this would mean that we lose functionality, and also makes vc
archaeology more difficult at the same time, so it's a lose-lose
proposition in my eyes. 

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: Eli Zaretskii <eliz@gnu.org>,  emacs-devel@gnu.org
> Date: Mon, 13 Sep 2021 14:05:26 +0200
> 
> And as a practical matter, a quick grep shows that we have more
> than 10K lines that match ";;.*`", so the code churn would be huge.

I do not intend to change all of them.  I wouldn't have changed those
few places, either, had Stefan Kangas not decide to fix the quoting
there.  If we are fixing quoting, we might as well do it right, or not
at all.

> So changing this would mean that we lose functionality, and also
> makes vc archaeology more difficult at the same time, so it's a
> lose-lose proposition in my eyes.

I didn't propose anything, I just changed a few lines that were
touched several hours before that anyway.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Lars Ingebrigtsen]

Eli Zaretskii <eliz@gnu.org> writes:

> I didn't propose anything, I just changed a few lines that were
> touched several hours before that anyway.

Oh, sorry -- I didn't notice that.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: master ff4de1b: Fix quoting style in Lisp comments

[Eli Zaretskii]

> Cc: emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Mon, 13 Sep 2021 14:51:32 +0300
> 
> > We decided to use this style in plain-text documentation long ago.
> 
> IME code comments are not the same as plain-text documentation.

Then what does?  The only place where we must leave `..' alone is in
doc strings, everything else is fair play.

> Aside from highlighting, we have other code that relies on unambiguous 
> markup. Emacs Lisp completion code also enters "symbol completion" mode 
> (so to speak) after you type '`', allowing one to more easily input an 
> existing symbol.

I didn't say anything about completion text.

> > Comments, NEWS, PROBLEMS, DEBUG, commit log messages (thus the
> > generated ChangeLog files), and any other form of plain-text
> > documentation.
> 
> Having docstring markup in plain comments is fairly common across 
> languages, I think.

Again, I don't see how this is relevant to the issue at hand.

Re: master ff4de1b: Fix quoting style in Lisp comments

[Dmitry Gutov]

On 13.09.2021 15:26, Eli Zaretskii wrote:
>> Cc: emacs-devel@gnu.org
>> From: Dmitry Gutov <dgutov@yandex.ru>
>> Date: Mon, 13 Sep 2021 14:51:32 +0300
>>
>>> We decided to use this style in plain-text documentation long ago.
>>
>> IME code comments are not the same as plain-text documentation.
> 
> Then what does?

I don't understand the question.

> The only place where we must leave `..' alone is in
> doc strings, everything else is fair play.

I think I have explained how changing comments this way breaks useful 
functionality.

>> Aside from highlighting, we have other code that relies on unambiguous
>> markup. Emacs Lisp completion code also enters "symbol completion" mode
>> (so to speak) after you type '`', allowing one to more easily input an
>> existing symbol.
> 
> I didn't say anything about completion text.
> 
>>> Comments, NEWS, PROBLEMS, DEBUG, commit log messages (thus the
>>> generated ChangeLog files), and any other form of plain-text
>>> documentation.
>>
>> Having docstring markup in plain comments is fairly common across
>> languages, I think.
> 
> Again, I don't see how this is relevant to the issue at hand.

References in docstrings ~ references in comments. With the same syntax.

`...' is our docstring syntax for references.